Tag: TG_NEURAL

DSSM model, trained for reformulations, in the document uses relevant to the request of the proposal

DSSM model with early binding, trained on reforming and learned by ASR hypotheses of musical requests for Alice

A model trained on a PRS-Law PRS to predict BERT, trained on sinsig_ce with threshold value 0.5, using a chain of regions to the country

DSSM model with early binding, trained on reforming and learned on musical requests for Alice

A pool of logs is marked with BERT trained on Sinsig. DSSM model is trained on this pool using BaseregionChain

DSSM Prediction of the probability of URL + Title that there is no product on the page.

DSSM Prediction of the probability of URL + Title, which is on the page one product.

DSSM Prediction of the probability of URL + Title, that there are a lot of goods on the page.

Finetuned reformulations DSSM to commercial clicked bargain odd-like target from visit log

DSSM Prediction of the probability of URL + Title that the document is an overlap.

Predict SOS.DSSM models by URL + Title.

Med.DSSM Predictions URL + Title models.

FIN_LAW.DSSM Predictions URL + TITLE.

Predict Cruelty.dssm URL + TITLE models.

Request and documentary navigation model.

Prediction of Med_with_Trash.DSSM (Medic. Document model with Tresh Valley in Lern) Models for URL + Title.

Prediction FIN_LAW_WITH_TRASH.DSSM (Fin-Jur. Document model with a tresh valve in Lern) Models for URL + Title.

DSSM, predicting the Page Quality rating for the document

DSSM CLOSE DISCOUNT according to data specific for Alice

The prediction of the total time spent to the end of the session, provided that this pair is implemented by the request-document

The prediction of the contribution of this pair request-document to the timetable

The prediction of the percentage of the length of the track, which will be lost subject to the implementation of this pair of the request

DSSM model trained on clicks.

DSSM model trained on clicks.

DSSM model trained on clicks.

DSSM model trained on clicks.

DSSM model trained on clicks.

DSSM model trained on clicks.

DSSM model trained on clicks.

DSSM model trained on clicks.

Neural model of content quality for medical subjects

Neural model of content quality for medical subjects (for ex -)

Neural model of content quality for financial and legal topics

Neural model of content quality for financial and legal topics (for exposures)

Neural model of content quality for SOS topics

Neural model of content quality for SOS subjects (for ex -)

Prediction of the time of the session, provided that this pair is requested by the request-document

Request-document DSSM, predicting the dog's dog

Request-document model Sinsiga.

DSSM, predicting whether the site is a facial

DSSM model trained on clicks. Takes bigrams into account.

DSSM model trained on clicks. Takes bigrams into account. Embeddings for documents are computed offline.

A minimum of gradients according to the Bigramm LogdwellTime model.

Maximum from gradients according to the Bigramm Logdwelltime model.

The second central moment (dispersion) from gradients according to the Bigramm Logdwelltime model.

The third central moment from gradients according to the Bigramm Logdwelltime model.

The probability that the VK.com host is popular for this request in accordance with the corresponding DSSM model.

The likelihood that the Onliner.by host is popular for this request according to the corresponding DSSM model.

The probability that the Rambler.ru host is popular for this request in accordance with the corresponding DSSM model.

The likelihood that the ExpertCen.ru host is popular for this request in accordance with the corresponding DSSM model.

The probability that the Sunhome.ru host is popular for this request according to the corresponding DSSM model.

DSSM model trained on clicks, target=OneClicks/Clicks. Takes bigrams into account.

DSSM model trained on clicks, target=QueryDwellTime stream value. Takes bigrams into account.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN (closest to the DSSM models trained to predict XFDTSHOW of extension). Aggregation on all extensions. The greatest balanced value of the factor. A mixture of many streams, weight is calculated by a fixed Polita component from the scales on this annotation. The algorithm for aggregation of words weights is BCLMMIXPLAIN: a linear mixture of annotation BCLM weights and balanced Positionless weights of the word, then the former meters are aggregated through BM15. Normalization coefficient 10^(-5).

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN (closest to the DSSM models trained to predict XFDTSHOW of extension). Aggregation on all extensions. The greatest balanced value of the factor. It is normalized for the maximum weight of expansion. Stream: Correctedctrlongperiod. The degree of coating of words query accurate to form (without synonyms).

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN (closest to the DSSM models trained to predict XFDTSHOW of extension). Aggregation on all extensions. The greatest balanced value of the factor. It is normalized for the maximum weight of expansion. Stream: Correctedctrlongperiod. The BCLMPLANEPROXIMITY15W0SIZE1 algorithm: uses BCLM with free weight if there are several words of the request, if the word is one, then the sum of hits is used like a coincidence. Normalization coefficient 0.001.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN (closest to the DSSM models trained to predict XFDTSHOW of extension). Aggregation on all extensions. The average weight of extensions.

Document DSSM model Language Classifier Rus.

Document DSSM model Language Classifier Eng.

Document DSSM model Language Classifier Other.

DSSM Prediction to determine Alice's irrelevant answers

Query-MainContentKeywords similarity, target: logDwellTime

Dssm Boosting query self similarity for XfWeight model.

Dssm Boosting AvgTop02Score aggregation for XfWeight model over 5-means centroids.

Dssm Boosting AvgTop04Score aggregation for XfWeight model over 5-means centroids.

Dssm Boosting AvgTop02ScoreAvgClusterTop3Weighted aggregation for XfWeight model over 5-means centroids.

Dssm Boosting AvgTop02Score aggregation for XfWeight model over 5-means centroids (query as expansion).

Dssm Boosting AvgTop02ScoreAvgClusterTop3Weighted aggregation for XfWeight model over 5-means centroids (query as expansion).

Dssm Boosting query self similarity for XfOne model.

Dssm Boosting Score aggregation for XfOne model over 1-means centroids.

Dssm Boosting ScaledSumWeight aggregation for XfOne model over 1-means centroids.

Dssm Boosting Score aggregation for XfOne model over 1-means centroids (query as expansion).

Dssm Boosting ScoreAvgNearest1Weighted aggregation for XfOne model over 1-means centroids (query as expansion).

Dssm Boosting ScoreAvgNearest5Weighted aggregation for XfOne model over 1-means centroids (query as expansion).

Dssm Boosting Score aggregation for XfOneSe model over 1-means centroids.

Dssm Boosting ScoreScaledSumWeighted aggregation for XfOneSe model over 1-means centroids.

Dssm Boosting ScoreAvgNearest5Weighted aggregation for XfOneSe model over 1-means centroids.

Dssm Boosting query self similarity for Ctr model.

Dssm Boosting Score aggregation for Ctr model over 1-means centroids.

Dssm Boosting Score aggregation for Ctr model over 1-means centroids (query as expansion).

Dssm Boosting ScoreScaledSumWeighted aggregation for Ctr model over 1-means centroids (query as expansion).

Dssm Boosting ScoreAvgNearest1Weighted aggregation for Ctr model over 1-means centroids (query as expansion).

DSSM model trained on CTRs without miner.

DSSM prediction (URL + Title), trained for the Page_QUALYY signal and implemented in RTHUB, the first slot.

DSSM prediction (URL + Title), trained on the Page_QUALYY signal and implemented in RTHUB, the second slot.

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

DSSM model trained on click odd pool

DSSM model trained on click personalization pool

DSSM model trained on click triangle pool

DSSM model trained on clicks without miner (with no-clicks and AM-hard negatives). Takes bigrams into account.

Predicted by demand and country, using a DSSM model, the length of the click from this country.

The average for the year for the year predicted using the neural network.

ADDTIME ADDTIME is predicted using a neural network for a year.

The average Txthirelesy value predicted using a neural network for the year.

The average Textlike is predicted using a neural network for the year.

The average HasnoallwordStrsy is predicted using a neural network for a year.

The average ISFORUM is predicted using a neural network for the year.

The average Haspayments is predicted using a neural network for the year.

The average value of Yabarhostavgtime2 for the year for the year.

The average yabarurlvisitors is predicted using a neural network for the year.

The average value of QueryDowneronlyClickRate for the year for the year.

The average Dateraage value for the year for a year predicted using a neural network.

The average LonGestText is predicted using a neural network for the year.

The average DifferentinTernallinks for the year for the year.

The average value of QueryDowneronlyClickRate_Rreg is predicted using a neural network for a year.

The average BOCM value predicted using a neural network for the year.

The average ISindEXPAGE is predicted using a neural network for the year.

The average value of QueriesavGCM2 for the year for the year predicted using a neural network.

The average BrowserhostdowLoadProbabolyity for the year for the year.

The average value of Regbrowseruserhub for the year for a year predicted using a neural network.

The average AuxtitlebM25 average value for the year for the year.

The average value of QuryurlCorrededCTRXFACTOR for the year for the year.

The average value of QueryTodoCallsumfcountTextbM11Norm16384 for the year for the year.

The average value of the XFDTSHOWALSUMWFSUMWBODYMINWINDOWSIZE for the year for the year.

The value of the ISMAINPAGE with clicks predicted using the neural network with clicks on request for the year.

A mid Yabarurlavgtime value predicted using a neural network with clicks for a year.

DiffferentinTernallinks, which is predicted using a neural network, is a weighted net with clicks for a year.

The Malue Network DwellTime-AMI predicted using the neural network is the value of Urldomainfraction for the year.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: BCLMWEIGHTEDFLOGW0 in the Stream group 3. The maximum balanced value of the factor.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: BM15FLOG in the group of streams 2. The maximum balanced value of the factor.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: ORIGINALREQUENTFRACTFRACTION OF THE FIELDSETBAGOFWORDS Stream.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: MixmatchweightedValue by Stream Querydwelltime. The maximum balanced value of the factor is normalized for the total weight.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: BM15 according to Stream Title. The total balanced values ​​of the factor multiplied by weight (\ frac {\ sum w_i * (w_i * f_i)} {\ sum w_i}) normalized for total weight.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: BCLMWEIGHTEDFLOGW0 in the Stream group 3. The minimum value of the factor for the expansion top.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: BCLMWEIGHTEDFLOGW0 in the Stream group 3. The total balanced values ​​of the factor multiplied by weight (\ frac {\ sum w_i * (w_i * f_i)} {\ sum w_i}) normalized for the total weight.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: BM15FLOG in the Stream group 1. The maximum balanced value of the factor.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: BM15FLOG in the Stream group 1. The total balanced value of the factor is normalized for the total weight.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: Bag AnnotationMatChavgvalue by Stream LongClicksp.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: BM15FLOG for the Stream group 1. The total balanced values ​​of the factor multiplied by weight (\ frac {\ sum w_i * (w_i * f_i)} {\ sum w_i}) for expansion top extensions, standardized for the total weight of the expansion top.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: BM15FLOG in the Stream group 1. The minimum balanced value of the factor for the expansion top extensions normalized for the maximum weight by the expansion top.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: PairminProximity according to Stream Body. The maximum balanced value of the factor is normalized for the total weight.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: BM15FLOG for the Stream group 1. The total balanced values ​​of the factor multiplied by weight (\ frac {\ sum w_i * (w_i * f_i)} {\ sum w_i}) normalized for total weight.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: SIMPLECLIC SIMPLECLICS bag.

Linguistic boosting factor. Type of extensions: XFDTSHOWKNN. Factor: CosinemaxMatch bag according to Title Stream.

The prediction of the probability that the request is localized in accordance with the regionality5 rule.

DSSM model trained on clicks without miner (with no-clicks and am_hard negatives 50/50 and then on am_hard negatives only). Takes bigrams into account.

Dssm Boosting Score aggregation for XfOneSeAmSsHard model over 1-means centroids.

Dssm Boosting ScoreAvgClusterTop3Weighted aggregation for XfOneSeAmSsHard model over 1-means centroids.

Linguistic boosting factor. Type of extensions: Querytotextbyxfdtshowknn. Factor: Norm256 by stream BOCM11. The total balanced values ​​of the factor multiplied by weight (\ frac {\ sum w_i * (w_i * f_i)} {\ sum w_i}).

Linguistic boosting factor. Type of extensions: Querytotextbyxfdtshowknn. Factor: Minwindowsize by Stream Body. The total balanced values ​​of the factor multiplied by weight (\ frac {\ sum w_i * (w_i * f_i)} {\ sum w_i}) by the expansion top, normalized for the total weight according to the expansion top.

Linguistic boosting factor. Type of extensions: Querytotextbyxfdtshowknn. Factor: Minwindowsize by Stream Body. The total balanced values ​​of the factor multiplied by weight (\ frac {\ sum w_i * (w_i * f_i)} {\ sum w_i}) normalized for total weight.

Linguistic boosting factor. Type of extensions: Querytotextbyxfdtshowknn. Factor: Norm256 by stream BOCM11. The total balanced values ​​of the factor multiplied by weight (\ frac {\ sum w_i * (w_i * f_i)} {\ sum w_i}) according to the expansion top.

Linguistic boosting factor. Type of extensions: Querytotextbyxfdtshowknn. The minimum expansion weight.

Linguistic boosting factor. Type of extensions: Querytotextbyxfdtshowknn. The arithmetic mean of expansion weights.

Linguistic boosting factor. Type of extensions: Querytotextbyxfdtshowknn. The total weight of the extensions.

Linguistic boosting factor. Type of extensions: Querytotextbyxfdtshowknn. Factor: ORIGINALREQUENTFRACTFRACTION OF THE FIELDSETBAGOFWORDS Stream.

Characterizes the request for the degree of change from the addition of a fixed word (number of some year), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

Characterizes a request for the degree of change from the addition of a fixed word ('online' for Kirilitsa), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

Characterizes the request for the degree of change from removing a fixed word ('site' for Kirilitsa), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

Neural document model for finding unexpected tin

DSSM prediction of google specificity for query

The average value of Randomlogqueryavgaddtime of the closest KNN queries.

The average value of RandomlogqueryavgtXthirelsy nearest KNN queries.

The average value of Randomlogqueryavgtextlike nearest KNN queries.

The average value of Randomlogqueryavgisforum of the closest KNN queries.

The average value of Randomlogqueryavghaspayments closest to KNN queries.

The average value of Randomlogqueryavgdiferentinternallinks of the nearest KNN queries.

The average value of RandomlogqueryavgistargetbussinessCard of the nearest KNN queries.

The average value is RandomlogqueryavgquerytododoCallsumfcountTextBM11NORM16384 of the nearest KNN queries.

The average value is Randomlogqueryavgxfdtshowallsumwfsumwbodyminwindowsize closest KNN queries.

The average value for the query factor according to Querytotextbyxfdtshowknn linguosting data is calculated in the LingBoostqueryFeatures Hotemail Rules

Quantile 0.1 for the request factor according to the XFDTSHOWKNN linguosting data, is calculated in the LingBoostqueryFeatures Hotemaway Rules

Quantile 0.9 for the quantity factor according to the XFDTSHOWKNN linguosting data, is calculated in the LingBoostqueryFeatures Hotemail Rules

Dssm Boosting Score for SerpSimilarityHard model over 1-means centroids.

The result of the use of a neural model, trained to distinguish long clicks from other events, the input of the model is the ambassadors and bigram meters, calculated by text streams (Title, Body, URL).

Linguistic boosting factor. Type of extensions: QFUFFILTEDBYXFONSE (QFUF, filtered on the DSSM models Xfonese). Aggregation on all extensions. The greatest value of the factor. Into aircraft association of the URLs, Title, Body, Correctedctr, Longclick, OneClick, Browserpagerank, Splitdwelltime, SampleperiodDayFrc, SimpleClick, Yabarvisits, Yabartime. The algorithm for aggregation of the scales of words: BM15FLOG (BM15 Aggregation of Logarithm of Construction of Words). Normalization coefficient 0.001.

Linguistic boosting factor. Type of extensions: QFUFFILTEDBYXFONSE (QFUF, filtered on the DSSM models Xfonese). Aggregation on all extensions. The greatest value of the factor. Rebelled association of streams Title, Body, LongClick, LongClicksp, OneClick. The algorithm for aggregation of the scales of words: BCLMWEIGHTEDFLOGW0. Normalization coefficient 0.001.

Linguistic boosting factor. Type of extensions: QFUFFILTEDBYXFONSE (QFUF, filtered on the DSSM models Xfonese). Aggregation on all extensions. The greatest value of the factor. It is considered to be composational stream, consisting of an tokenized Url and a title of a document. The algorithm for aggregation of the scales of words: BM15FLOGW0. Normalization coefficient 0.0001.

Linguistic boosting factor. Type of extensions: QFUFFILTEDBYXFONSE (QFUF, filtered on the DSSM models Xfonese). Aggregation on all extensions. The greatest value of the factor. It is considered according to the heading of the document. The algorithm for aggregation of the scales of words: BM15. Normalization coefficient 0.1.

Linguistic boosting factor. Type of extensions: QFUFFILTEDBYXFONSE (QFUF, filtered on the DSSM models Xfonese). Aggregation by TOP-10 (by the value of the factor) extensions. A suspended sum of the Libra of factors. Normalized for the total weight of extensions. Into aircraft association of the URLs, Title, Body, Correctedctr, Longclick, OneClick, Browserpagerank, Splitdwelltime, SampleperiodDayFrc, SimpleClick, Yabarvisits, Yabartime. The algorithm for aggregation of the scales of words: BM15FLOG (BM15 Aggregation of Logarithm of Construction of Words). Normalization coefficient 0.001.

Linguistic boosting factor. Type of extensions: QFUFFILTEDBYXFONSE (QFUF, filtered on the DSSM models Xfonese). Aggregation by TOP-10 (by the value of the factor) extensions. A suspended sum of the Libra of factors. Normalized for the total weight of extensions. It is considered according to the contents of the document. The minimum window size, which includes all the words of the request. It is normalized for the number of words in the request.

The factor for the filtered original request: the DSSM state from the request is calculated without words to the initial request, after which the threshold is cut off. Into aircraft association of the URLs, Title, Body, Links, Correctedctr, LongClick, OneClick, Browserpagerank, Splitdwelltime, SampleperiodDayFrc, SimpleClick, Yabarvisits, Yabartime. The algorithm for aggregation of the scales of words: BM15FLOG (BM15 Aggregation of Logarithm of Construction of Words). Normalization coefficient 0.001.

The factor for the filtered original request: the DSSM state from the request is calculated without words to the initial request, after which the threshold is cut off. It is considered to be composational stream, consisting of an tokenized Url and a title of a document. The algorithm for aggregation of the scales of words: BM15FLOGW0. Normalization coefficient 0.0001.

DSSM model trained on cross language CTRs using serp similarity hard miner.

For all words of the request, the weight is calculated by the Query-Mutation method (the distance between the requests in nash and there is no word). The sum of the scales of the words found in the title is taken, divided by the sum of the scales of all words.

For all words of the request, the weight is calculated by the Query-Mutation method (the distance between the requests in nash and there is no word). Maximum weight is taken among words absent in the title of the document.

The result of the use of a neural model, trained to distinguish long clicks from other events, the input of the model is the ambassadors and bigram meters calculated by text streams (Body, URL).

Linguistic boosting factor. Type of extensions: XFONESEKNN (closest to the DSSM models trained to predict XFDTSHOW of extension). Aggregation on all extensions. The greatest balanced value of the factor. It is normalized for the maximum weight of expansion. Into aircraft association of the URLs, Title, Body, Links, Correctedctr, LongClick, OneClick, Browserpagerank, Splitdwelltime, SampleperiodDayFrc, SimpleClick, Yabarvisits, Yabartime. The algorithm for aggregation of the scales of words: BM15FLOG (BM15 Aggregation of Logarithm of Construction of Words). Normalization coefficient 0.001.

Linguistic boosting factor. Type of extensions: XFONESEKNN (closest to the DSSM models trained to predict XFDTSHOW of extension). Aggregation on all extensions. The greatest balanced value of the factor. It is normalized for the maximum weight of expansion. Todo Algorithm: The maximum weight of the completely coincided with the request of the annotation. It is considered according to Stream OneClick.

Linguistic boosting factor. Type of extensions: QuerytotextByxfoneKnn (Querytotext extensions of Xfoneeseknn extensions). Aggregation by TOP-10 (by the value of the factor) extensions. A suspended sum of the Libra of factors. Normalized for the total weight of extensions. It is considered according to the contents of the document. The minimum window size, which includes all the words of the request. It is normalized for the number of words in the request.

Linguistic boosting factor. Type of extensions: QuerytotextByxfoneKnn (Querytotext extensions of Xfoneeseknn extensions). Aggregation on all extensions. A suspended sum of the Libra of factors. Normalized for the total weight of extensions. Rebelled association of streams Title, Body, LongClick, LongClicksp, OneClick. The algorithm for aggregation of the scales of words: BCLMWEIGHTEDFLOGW0. Normalization coefficient 0.001.

DSSM model that predicts the logarithm of the longest click on the Serpa. As negative examples, select Urla from past requests of the same user, and the maximum time between requests is no more than 7 minutes (super -cords for reformulations)

DSSM model with early binding, trained in reformulations, which predicts the logarithm of the longest click on the Serpa.

Neural network value for contexts of query hits in document text. Predicts relevance-all-8-years. Uses formula ussr-dump-20190719 prs-20190720 all-8-years [t > 0.25] CrossEntropy 20k 0.25 -S 0.8 -Z 1 predictions for learning.

DSSM model trained on a reformal pool, which in the request, in addition to the request itself, receives 4 extensions of the XFDT with the largest weight

A model trained to predict an assessment of the USSR-DUMP-20190719 PRS-20190720 ALL-8-YEARS [T> 0.25] Crossentropy 20K 0.25 -s 0.8 -z 1.

Neuron document model for finding unexpected tin (for ex -)

A model trained to predict an assessment of the USSR-DUMP-20190719 PRS-20190720 ALL-8-YEARS [T> 0.25] Crossentropy 20K 0.25 -s 0.8 -z 1 and an educational study on assessments of relevance.

Word LSTM trained on quasar toloka data (alarms_timers prediction)

Word LSTM trained on quasar toloka data (commands prediction)

Word LSTM trained on quasar toloka data (fairy_tails prediction)

Word LSTM trained on quasar toloka data (games_skills prediction)

Word LSTM trained on quasar toloka data (geo prediction)

Word LSTM trained on quasar toloka data (info_request prediction)

Word LSTM trained on quasar toloka data (iot prediction)

Word LSTM trained on quasar toloka data (market prediction)

Word LSTM trained on quasar toloka data (music prediction)

Word LSTM trained on quasar toloka data (new_functionality prediction)

Word LSTM trained on quasar toloka data (news prediction)

Word LSTM trained on quasar toloka data (ontofacts prediction)

Word LSTM trained on quasar toloka data (other prediction)

Word LSTM trained on quasar toloka data (radio prediction)

Word LSTM trained on quasar toloka data (search prediction)

Word LSTM trained on quasar toloka data (taxi prediction)

Word LSTM trained on quasar toloka data (tell_something prediction)

Word LSTM trained on quasar toloka data (time_tables prediction)

Word LSTM trained on quasar toloka data (toloka_gc prediction)

Word LSTM trained on quasar toloka data (translate prediction)

Word LSTM trained on quasar toloka data (video prediction)

Word LSTM trained on quasar toloka data (weather prediction)

Word LSTM trained on quasar toloka data (alarms_timers prediction)

Word LSTM trained on quasar toloka data (commands prediction)

Word LSTM trained on quasar toloka data (fairy_tails prediction)

Word LSTM trained on quasar toloka data (games_skills prediction)

Word LSTM trained on quasar toloka data (geo prediction)

Word LSTM trained on quasar toloka data (info_request prediction)

Word LSTM trained on quasar toloka data (iot prediction)

Word LSTM trained on quasar toloka data (market prediction)

Word LSTM trained on quasar toloka data (music prediction)

Word LSTM trained on quasar toloka data (new_functionality prediction)

Word LSTM trained on quasar toloka data (news prediction)

Word LSTM trained on quasar toloka data (ontofacts prediction)

Word LSTM trained on quasar toloka data (other prediction)

Word LSTM trained on quasar toloka data (player_commands prediction)

Word LSTM trained on quasar toloka data (radio prediction)

Word LSTM trained on quasar toloka data (sound_commands prediction)

Word LSTM trained on quasar toloka data (taxi prediction)

Word LSTM trained on quasar toloka data (tell_joke prediction)

Word LSTM trained on quasar toloka data (tell_something prediction)

Word LSTM trained on quasar toloka data (toloka_gc prediction)

Word LSTM trained on quasar toloka data (translate prediction)

Word LSTM trained on quasar toloka data (tv_intents prediction)

Word LSTM trained on quasar toloka data (video prediction)

Word LSTM trained on quasar toloka data (vins_intents prediction)

Word LSTM trained on quasar toloka data (weather prediction)

Word LSTM traind on toloka data (music_play prediction)

Word LSTM traind on toloka data (video_play prediction)

Word LSTM traind on scenarios (VINS) data (music_play prediction)

Word LSTM traind on scenarios (VINS) data (video_play prediction)

Word LSTM traind on toloka data (search prediction)

Word LSTM traind on scenarios (VINS) data (search prediction)

Word LSTM trained on toloka data (music_fairy_tale prediction)

Word LSTM trained on scenarios (VINS) data (music_fairy_tale prediction)

Word LSTM trained on toloka data (music_ambient_sound prediction)

Word LSTM trained on scenarios (VINS) data (music_ambient_sound prediction)

Word LSTM trained on toloka data (radio_play prediction)

Word LSTM trained on scenarios (VINS) data (radio_play prediction)

Word LSTM trained on scenarios (VINS) data (music_sing_song prediction)

Word LSTM trained on toloka data (music_sing_song prediction)

Word LSTM trained on scenarios (VINS) data (find_poi prediction)

Word LSTM trained on toloka data (find_poi prediction)

Word LSTM trained on scenarios (VINS) data (open_site_or_app prediction)

Word LSTM trained on toloka data (open_site_or_app prediction)

Word LSTM trained on scenarios (VINS) data (how_much prediction)

Word LSTM trained on toloka data (how_much prediction)

Word LSTM trained on scenarios (VINS) data (translate prediction)

Word LSTM trained on toloka data (translate prediction)

Word LSTM trained on scenarios (VINS) data (tv_broadcast prediction)

Word LSTM trained on toloka data (tv_broadcast prediction)

Word LSTM trained on scenarios (VINS) data (convert prediction)

Word LSTM trained on toloka data (convert prediction)

Word LSTM trained on scenarios (VINS) data (player_dislike prediction)

Word LSTM trained on scenarios (VINS) data (create_reminder prediction)

Word LSTM trained on toloka data (create_reminder prediction)

Word LSTM trained on scenarios (VINS) data (tv_stream prediction)

Word LSTM trained on toloka data (tv_stream prediction)

Word LSTM trained on scenarios (VINS) data (recite_a_poem prediction)

Word LSTM trained on toloka data (recite_a_poem prediction)

Gc DSSM classifier distilled from BERT

Word LSTM trained on factoid markup from toloka

The probability that the VK.com host is popular for this request in accordance with the corresponding DSSM model.

The probability that the Onliner.by host is popular for this request according to the corresponding DSSM model.

The probability that the Rambler.ru host is popular for this request in accordance with the corresponding DSSM model.

The likelihood that the ExpertCen.ru host is popular for this request in accordance with the corresponding DSSM model.

The probability that the Sunhome.ru host is popular for this request according to the corresponding DSSM model.

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

Predicted by demand and country, using a DSSM model, the length of the click from this country.

The average for the year for the year predicted using the neural network.

ADDTIME ADDTIME is predicted using a neural network for a year.

The average Txthirelesy value predicted using a neural network for the year.

The average Textlike is predicted using a neural network for the year.

The average HasnoallwordStrsy is predicted using a neural network for a year.

The average ISFORUM is predicted using a neural network for the year.

The average Haspayments is predicted using a neural network for the year.

The average value of Yabarhostavgtime2 for the year for the year.

The average yabarurlvisitors is predicted using a neural network for the year.

The average value of QueryDowneronlyClickRate for the year for the year.

The average Dateraage value for the year for a year predicted using a neural network.

The average LonGestText is predicted using a neural network for the year.

The average DifferentinTernallinks for the year for the year.

The average value of QueryDowneronlyClickRate_Rreg is predicted using a neural network for a year.

The average ISTARGETBUSSINESSCARD is predicted using a neural network for the year.

The average BOCM value predicted using a neural network for the year.

The average ISindEXPAGE is predicted using a neural network for the year.

The average value of QueriesavGCM2 for the year for the year predicted using a neural network.

The average BrowserhostdowLoadProbabolyti is predicted using a neural network for the year.

The average value of Regbrowseruserhub for the year for a year predicted using a neural network.

The average AuxtitlebM25 average value for the year for the year.

The average value of QueryurlCorrectrxFactor for the year for the year.

The average value of QueryTodoCallsumfcountTextbM11Norm16384 for the year for the year.

The average value of the XFDTSHOWALSUMWFSUMWBODYMINWINDOWSIZE for the year for the year.

The value of the ISMAINPAGE with clicks predicted using the neural network with clicks on request for the year.

A mid Yabarurlavgtime value predicted using a neural network with clicks for a year.

DiffferentinTernallinks, which is predicted using a neural network, is a weighted net with clicks for a year.

The Malue Network DwellTime-AMI predicted using the neural network is the value of Urldomainfraction for the year.

The prediction of the probability that the request is localized in accordance with the regionality5 rule.

Characterizes the request for the degree of change from removing a fixed word ('site' for Kirilitsa), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

Characterizes the request for the degree of change from the addition of a fixed word (number of some year), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

Characterizes a request for the degree of change from the addition of a fixed word ('online' for Kirilitsa), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

DSSM prediction of google specificity for query

The average value of Randomlogqueryavgaddtime of the closest KNN queries.

The average value of RandomlogqueryavgtXthirelsy nearest KNN queries.

The average value of Randomlogqueryavgtextlike nearest KNN queries.

The average value of Randomlogqueryavgisforum of the closest KNN queries.

The average value of Randomlogqueryavghaspayments closest to KNN queries.

The average value of Randomlogqueryavgdiferentinternallinks of the nearest KNN queries.

The average value of RandomlogqueryavgistargetbussinessCard of the nearest KNN queries.

The average value is RandomlogqueryavgquerytododoCallsumfcountTextbM11NORM16384 of the nearest KNN queries.

The average value is Randomlogqueryavgxfdtshowallsumwfsumwbodyminwindowsize closest KNN queries.

The average value for the query factor according to Querytotextbyxfdtshowknn linguosting data is calculated in the LingBoostqueryFeatures Hotemail Rules

Quantile 0.1 for the request factor according to the XFDTSHOWKNN linguosting data, is calculated in the LingBoostqueryFeatures Hotemaway Rules

Quantile 0.9 for the quantity factor according to the XFDTSHOWKNN linguosting data, is calculated in the LingBoostqueryFeatures Hotemaway Rules

The query getting into the TOP 11 clusters on Kernel metric based on DSSM proximity.

DSSM distance from request to word match

DSSM distance from request to word hockey

DSSM distance from request to word broadcast

DSSM distance from request to word championship

DSSM distance from request to word series

DSSM distance from request to word today

DSSM distance from request to word last

DSSM Distance from request to the word accident

DSSM distance from request to word news

DSSM distance from request to word death

DSSM distance from request to word football

DSSM distance from request to word day

DSSM distance from request to word postcard

DSSM distance from request to word khl

DSSM distance from request to word UFC

DSSM distance from request to word results

DSSM distance from request to word schedule

DSSM distance from request to word fresh

DSSM distance from request to word biography

DSSM distance from request to word yesterday

DSSM distance from request to word forecast

Requests got into the 34th cluster based on CTR-DSSM.

DSSM distance from request to word coronavirus

DSSM distance from request to word buy

DSSM distance from request to word price

DSSM distance from request to word product

DSSM distance from the request to the word description

DSSM distance from request to word delivery

DSSM distance from request to word is paid

DSSM distance from request to word analogue

DSSM distance from request to word sale

DSSM distance from the request to the word installment

DSSM distance from request to word rent

DSSM distance from request to word catalog

DSSM distance from request to word store

Dssm distance from request to word watch

DSSM distance from request to word movie

DSSM distance from request to word season

DSSM distance from request to word free

The likelihood that the selected hosts from the Construction category are popular for this request according to the corresponding DSSM model

The likelihood that the selected hosts from the Wear category are popular for this request according to the corresponding DSSM model

The likelihood that the selected hosth hosts category are popular for this request in accordance with the corresponding DSSM model

The likelihood that the selected hosts from the Food category are popular for this request in accordance with the corresponding DSSM model

The likelihood that the selected hosts from the Furniture category are popular for this request according to the corresponding DSSM model

The likelihood that the selected hosts from the Travel category are popular for this request in accordance with the corresponding DSSM model

The likelihood that the selected hostplace hostplace category are popular for this request according to the corresponding DSSM model

The likelihood that the selected hosts from the Book category are popular for this request in accordance with the corresponding DSSM model

The likelihood that the selected hosts from the Electron category are popular for this request in accordance with the corresponding DSSM model

The likelihood that the selected hosts from the Board category are popular for this request in accordance with the corresponding DSSM model

The likelihood that the selected hosts from the Service category are popular for this request in accordance with the corresponding DSSM model

The likelihood that the selected hosto hosto are popular for this request in accordance with the corresponding DSSM model

The likelihood that the selected hosts from the Cinema category are popular for this request in accordance with the corresponding DSSM model

DSSM distance from request to word treatment

DSSM distance from request to word pharmacy

DSSM distance from request to the word cold

DSSM distance from request to word diet

DSSM distance from request to word ointment

DSSM distance from request to word means

DSSM distance from request to word recipe

DSSM distance from request to word belly

DSSM Prediction of money spent on a request

Prediction of the classifier that the request is a medical

Prediction of the classifier that the request is a legal

Prediction of the classifier that the request is financial

The prediction of the classifier is that the request of SOS topics

The prediction of the classifier that the request is navigation

The prediction of the classifier is that the request is white (that is, it is not worth showing tin on it)

DSSM prediction of the complexity of the request

The prediction of the classifier is that upon request you need to show URL from a mobile market

Prediction of the classifier to especially commercial requests

Prediction of the CP Classifier Requests

Requested classifier of programmatic queries V2

Classifier of a request for equipment repair services: office, household, computers, phones

Request classifier to Pro-Hoogle Expertise.

Whether the request is porn.

Prediction of the CP Classifier Requests

Prediction of the requesting classifier of violence

Prediction of the requesting classifier of weapons or drugs

The prediction of the classifier is that a culinary request request

The prediction of the classifier is that the request is strong international and culinary

Classifier of a request for pornographic for Indonesian

Classifier of requests for the requirement only English -speaking pages

Classifier of requests for the requirement of mixed English -speaking and Russian -speaking pages

The classifier of requests for the requirement of either only English -speaking pages, or English -speaking and Russian -speaking pages together

Wizdeteration Detector on top of DSSM Multiclick

WizDeteration service request detector on top of DSSM Multiclick

Wizdeteration Detector on top of DSSM ReformulationslonGestClicklogdt

Wizdeteration Detector on top of DSSM Multiclick on the cuts of the Docdoc.ru service shows

Telemedical Request detector from WizDeteration on top of DSSM Multiclick

Wizdeteration Detector on top of DSSM Multiclick

WizDeteration service request detector on top of DSSM Multiclick

Wizdeteration Detector on top of DSSM ReformulationslonGestClicklogdt

Wizdeteration Detector on top of DSSM Multiclick on the cuts of the Docdoc.ru service shows

Telemedical Request detector from WizDeteration on top of DSSM Multiclick

The average DifferentinTernallinks for the year for the year.

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The average value of Regbrowseruserhub for the year for a year predicted using a neural network.

The Malue Network DwellTime-AMI predicted using the neural network is the value of Urldomainfraction for the year.

The average Textlike is predicted using a neural network for the year.

Characterizes the request for the degree of change from the addition of a fixed word (number of some year), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

The average ISFORUM is predicted using a neural network for the year.

ADDTIME ADDTIME is predicted using a neural network for a year.

The average ISindEXPAGE is predicted using a neural network for the year.

The probability that the VK.com host is popular for this request in accordance with the corresponding DSSM model.

The probability that the Onliner.by host is popular for this request according to the corresponding DSSM model.

The probability that the Rambler.ru host is popular for this request in accordance with the corresponding DSSM model.

The likelihood that the ExpertCen.ru host is popular for this request in accordance with the corresponding DSSM model.

The probability that the Sunhome.ru host is popular for this request according to the corresponding DSSM model.

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

Predicted by demand and country, using a DSSM model, the length of the click from this country.

The average for the year for the year predicted using the neural network.

ADDTIME ADDTIME is predicted using a neural network for a year.

The average Txthirelesy value predicted using a neural network for the year.

The average Textlike is predicted using a neural network for the year.

The average HasnoallwordStrsy is predicted using a neural network for a year.

The average ISFORUM is predicted using a neural network for the year.

The average Haspayments is predicted using a neural network for the year.

The average value of Yabarhostavgtime2 for the year for the year.

The average yabarurlvisitors is predicted using a neural network for the year.

The average value of QueryDowneronlyClickRate for the year for the year.

The average Dateraage value for the year for a year predicted using a neural network.

The average LonGestText is predicted using a neural network for the year.

The average DifferentinTernallinks for the year for the year.

The average value of QueryDowneronlyClickRate_Rreg is predicted using a neural network for a year.

The average ISTARGETBUSSINESSCARD is predicted using a neural network for the year.

The average BOCM value predicted using a neural network for the year.

The average ISindEXPAGE is predicted using a neural network for the year.

The average value of QueriesavGCM2 for the year for the year predicted using a neural network.

The average BrowserhostdowLoadProbabolyti is predicted using a neural network for the year.

The average value of Regbrowseruserhub for the year for a year predicted using a neural network.

The average AuxtitlebM25 average value for the year for the year.

The average value of QueryurlCorrectrxFactor for the year for the year.

The average value of QueryTodoCallsumfcountTextbM11Norm16384 for the year for the year.

The average value of the XFDTSHOWALSUMWFSUMWBODYMINWINDOWSIZE for the year for the year.

The value of the ISMAINPAGE with clicks predicted using the neural network with clicks on request for the year.

A mid Yabarurlavgtime value predicted using a neural network with clicks for a year.

DiffferentinTernallinks, which is predicted using a neural network, is a weighted net with clicks for a year.

The Malue Network DwellTime-AMI predicted using the neural network is the value of Urldomainfraction for the year.

The prediction of the probability that the request is localized in accordance with the regionality5 rule.

Characterizes the request for the degree of change from removing a fixed word ('site' for Kirilitsa), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

Characterizes the request for the degree of change from the addition of a fixed word (number of some year), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

Characterizes a request for the degree of change from the addition of a fixed word ('online' for Kirilitsa), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

DSSM prediction of google specificity for query

The average value of Randomlogqueryavgaddtime of the closest KNN queries.

The average value of RandomlogqueryavgtXthirelsy nearest KNN queries.

The average value of Randomlogqueryavgtextlike nearest KNN queries.

The average value of Randomlogqueryavgisforum of the closest KNN queries.

The average value of Randomlogqueryavghaspayments closest to KNN queries.

The average value of Randomlogqueryavgdiferentinternallinks of the nearest KNN queries.

The average value of RandomlogqueryavgistargetbussinessCard of the nearest KNN queries.

The average value is RandomlogqueryavgquerytododoCallsumfcountTextbM11NORM16384 of the nearest KNN queries.

The average value is Randomlogqueryavgxfdtshowallsumwfsumwbodyminwindowsize closest KNN queries.

The average value of the request factor according to lingvobusting - Querytotextbyxfdtshowknn, is calculated in the Hippo rule LingBoostqueryFeatures

Quantile 0.1 for the request factor according to the XFDTSHOWKNN linguosting data, is calculated in the LingBoostqueryFeatures Hotemaway Rules

Quantile 0.9 for the quantity factor according to the XFDTSHOWKNN linguosting data, is calculated in the LingBoostqueryFeatures Hotemaway Rules

The average DifferentinTernallinks for the year for the year.

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The average value of Regbrowseruserhub for the year for a year predicted using a neural network.

The Malue Network DwellTime-AMI predicted using the neural network is the value of Urldomainfraction for the year.

The average Textlike is predicted using a neural network for the year.

Characterizes the request for the degree of change from the addition of a fixed word (number of some year), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

The average ISFORUM is predicted using a neural network for the year.

ADDTIME ADDTIME is predicted using a neural network for a year.

The average ISindEXPAGE is predicted using a neural network for the year.

Max cosine similarity between document and user history with clicks dwelltime > 30sec, by realtime user_history

Max cosine similarity between document and user history with clicks dwelltime > 60sec, by realtime user_history

Max cosine similarity between document and user history with clicks dwelltime > 120sec, by realtime user_history

Max cosine similarity between document and user history with clicks dwelltime > 180sec, by realtime user_history

Max cosine similarity between document and user history with clicks dwelltime > 360sec, by realtime user_history

Max cosine similarity between document and user history with clicks dwelltime <= 30sec, by realtime user_history

Max cosine similarity between document and user history with clicks dwelltime <= 60sec, by realtime user_history

Max cosine similarity between document and user history with clicks dwelltime <= 120sec, by realtime user_history

Max cosine similarity between document and user history with clicks dwelltime <= 180sec, by realtime user_history

Max cosine similarity between document and user history with clicks dwelltime <= 360sec, by realtime user_history

Avg by top-5 maximum cosine similarity between document and user history with clicks dwelltime > 30sec, by realtime user_history

Avg by top-5 maximum cosine similarity between document and user history with clicks dwelltime > 60sec, by realtime user_history

Avg by top-5 maximum cosine similarity between document and user history with clicks dwelltime > 120sec, by realtime user_history

Avg by top-5 maximum cosine similarity between document and user history with clicks dwelltime > 180sec, by realtime user_history

Avg by top-5 maximum cosine similarity between document and user history with clicks dwelltime > 360sec, by realtime user_history

Avg by top-10 maximum cosine similarity between document and user history with clicks dwelltime > 30sec, by realtime user_history

Avg by top-10 maximum cosine similarity between document and user history with clicks dwelltime > 60sec, by realtime user_history

Avg by top-10 maximum cosine similarity between document and user history with clicks dwelltime > 120sec, by realtime user_history

Avg by top-10 maximum cosine similarity between document and user history with clicks dwelltime > 180sec, by realtime user_history

Avg by top-10 maximum cosine similarity between document and user history with clicks dwelltime > 360sec, by realtime user_history

Avg by top-15 maximum cosine similarity between document and user history with clicks dwelltime > 30sec, by realtime user_history

Avg by top-15 maximum cosine similarity between document and user history with clicks dwelltime > 60sec, by realtime user_history

Avg by top-15 maximum cosine similarity between document and user history with clicks dwelltime > 120sec, by realtime user_history

Avg by top-15 maximum cosine similarity between document and user history with clicks dwelltime > 180sec, by realtime user_history

Avg by top-15 maximum cosine similarity between document and user history with clicks dwelltime > 360sec, by realtime user_history

The probability that the VK.com host is popular for this request in accordance with the corresponding DSSM model.

The probability that the Onliner.by host is popular for this request according to the corresponding DSSM model.

The probability that the Rambler.ru host is popular for this request in accordance with the corresponding DSSM model.

The probability that the Sunhome.ru host is popular for this request according to the corresponding DSSM model.

Predicted by demand and country, using a DSSM model, the length of the click from this country.

The average for the year for the year predicted using the neural network.

ADDTIME ADDTIME is predicted using a neural network for a year.

The average Txthirelesy value predicted using a neural network for the year.

The average Textlike is predicted using a neural network for the year.

The average HasnoallwordStrsy is predicted using a neural network for a year.

The average ISFORUM is predicted using a neural network for the year.

The average value of Yabarhostavgtime2 for the year for the year.

The average yabarurlvisitors is predicted using a neural network for the year.

The average value of QueryDowneronlyClickRate for the year for the year.

The average Dateraage value for the year for a year predicted using a neural network.

The average LonGestText is predicted using a neural network for the year.

The average DifferentinTernallinks for the year for the year.

The average value of QueryDowneronlyClickRate_Rreg is predicted using a neural network for a year.

The average ISTARGETBUSSINESSCARD is predicted using a neural network for the year.

The average BOCM value predicted using a neural network for the year.

The average ISindEXPAGE is predicted using a neural network for the year.

The average value of QueriesavGCM2 for the year for the year predicted using a neural network.

The average BrowserhostdowLoadProbabolyti is predicted using a neural network for the year.

The average value of Regbrowseruserhub for the year for a year predicted using a neural network.

The average value of QueryurlCorrectrxFactor for the year for the year.

The average value of QueryTodoCallsumfcountTextbM11Norm16384 for the year for the year.

The average value of the XFDTSHOWALSUMWFSUMWBODYMINWINDOWSIZE for the year for the year.

The value of the ISMAINPAGE with clicks predicted using the neural network with clicks on request for the year.

A mid Yabarurlavgtime value predicted using a neural network with clicks for a year.

DiffferentinTernallinks, which is predicted using a neural network, is a weighted net with clicks for a year.

The Malue Network DwellTime-AMI predicted using the neural network is the value of Urldomainfraction for the year.

Characterizes the request for the degree of change from the addition of a fixed word (number of some year), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

It characterizes the request for the degree of change from the addition of a fixed word ('online' for Kirilitsa), the DSSM model DSSMBOOSTINGXFONESEAMSHARD is used. It is also used in the ranking of ether.

Characterizes the request for the degree of change from removing a fixed word ('site' for Kirilitsa), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

DSSM prediction of google specificity for query

The maximum cosine of the current document with documents from history closed shorter than 30 seconds

The maximum cosine of the current document with documents from history, clicks longer than 180 seconds

The maximum cosine of the current document with documents from history, clicks longer than 30 seconds

dot product of DocDssmLogDwellTimeBigramsL3Embedding for left and right urls

RecDssmSpyTitleDomainEmb12Dot

RecCFSharpDomainDot

The result of the prediction of the DSCMCA above the features, Itditp-1084

A model that is according to Hesham Urlov/hosts from the history of the user and the URL, the Title document predicts the user's click in the recommendations of the BC widget, Middle-268

A model that is according to Hesham Urlov/hosts from the history of the user and the URL, the Title document predicts the user's click in the Zen Cards from the Z-Grouping, Itditp-1139

A model that is according to Hesham Urlov/hosts from the history of the user and the URL, the Title document predicts the user's click into arbitrary hubborn cards

A model that, according to Hesham Urlov/hosts from the history of the user (dedicated part) and the toned daily embeds (constant part) predicts a click

A model that, according to Hesham Urlov/hosts from the history of the user (dedicated part) and toned dock emobedam (constant part) predicts a click. Trained on Target Turbbo recommendations

DSSM model trained on clicks. Takes bigrams into account. Embeddings for documents are computed offline.

Query-MainContentKeywords similarity, target: logDwellTime

Dssm Boosting Score aggregation for XfOneSe model over 1-means centroids.

Dssm Boosting ScoreScaledSumWeighted aggregation for XfOneSe model over 1-means centroids.

Dssm Boosting ScoreAvgNearest5Weighted aggregation for XfOneSe model over 1-means centroids.

DSSM model of panther terms. Embed -expanded bar are used for L2. Top5 QFUF of Cosine proximity to the Embed document is filtered. As a factor, a minimum of the weight of the expansion of proximity is used.

DSSM model of panther terms. Embed -expanded bar are used for L2. Top5 QFUF of Cosine proximity to the Embed document is filtered. Next, the weighed amount of proximity is calculated, the weight of the expansion is used in the quality of weight, normalized for the amount of the weights of filtered extensions.

DSSM model of panther terms. Embed -expanded bar are used for L2. As a factor, the average proximity to the Embed of the document is used.

DSSM model of panther terms. Embed -expanded bar are used for L2. As a factor, a minimum of the weight of the expansion of proximity is used.

Meta:Resid metafactor on web_itditp:DssmLogDwellTimeBigramsDot(606)

SD:DFT_GREATER_CNT metafactor on web_itditp:DssmBoostingXfOneSeAmSsHardDot(657)

Meta:ResidMin metafactor on web_itditp:DssmLogDwellTimeBigramsDot(606)

MutualSerp:DFT_MAX metafactor on web_itditp:DssmBoostingXfWtd45Dot(632)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime less than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime less than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime more than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime less than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime more than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime less than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime less than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime more than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime less than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime more than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime less than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime less than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime more than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime less than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime more than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime less than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime less than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime more than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime less than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime more than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime less than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime less than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime more than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime less than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime more than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime less than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime less than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime more than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime less than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime more than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime less than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime less than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime more than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime less than 120sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime more than 120sec)

Cosine similarity between query embedding and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime less than 120sec)

Cosine similarity between query embedding and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 120sec)

Cosine similarity between query embedding and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime less than 120sec)

Cosine similarity between query embedding and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime more than 120sec)

Cosine similarity between query embedding and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime less than 120sec)

Cosine similarity between query embedding and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime more than 120sec)

Top1 weighted average cosine similarity between document and clustered embedding of documents from user_history (model=LogDwellTimeBigrams, dwelltime more than 120sec)

Top1 weighted average cosine similarity between document and clustered embedding of documents from user_history (model=LogDwellTimeBigrams, dwelltime less than 120sec)

Top2 weighted average cosine similarity between document and clustered embedding of documents from user_history (model=LogDwellTimeBigrams, dwelltime more than 120sec)

Top2 weighted average cosine similarity between document and clustered embedding of documents from user_history (model=LogDwellTimeBigrams, dwelltime less than 120sec)

cnt / (1 + cnt), cnt = sum of weights where cosine between document and clustered embedding of documents from user_history > 0.30 (model=LogDwellTimeBigrams, dwelltime more than 120sec)

cnt / (1 + cnt), cnt = sum of weights where cosine between document and clustered embedding of documents from user_history > 0.30 (model=LogDwellTimeBigrams, dwelltime less than 120sec)

cnt / (1 + cnt), cnt = sum of weights where cosine between document and clustered embedding of documents from user_history > 0.60 (model=LogDwellTimeBigrams, dwelltime more than 120sec)

cnt / (1 + cnt), cnt = sum of weights where cosine between document and clustered embedding of documents from user_history > 0.60 (model=LogDwellTimeBigrams, dwelltime less than 120sec)

cnt / (1 + cnt), cnt = sum of weights where cosine between document and clustered embedding of documents from user_history > 0.90 (model=LogDwellTimeBigrams, dwelltime more than 120sec)

cnt / (1 + cnt), cnt = sum of weights where cosine between document and clustered embedding of documents from user_history > 0.90 (model=LogDwellTimeBigrams, dwelltime less than 120sec)

Top1 weighted average cosine similarity between query embedding and clustered embedding of documents from user_history (model=LogDwellTimeBigrams, dwelltime more than 120sec)

Top1 weighted average cosine similarity between query embedding and clustered embedding of documents from user_history (model=LogDwellTimeBigrams, dwelltime less than 120sec)

Top2 weighted average cosine similarity between query embedding and clustered embedding of documents from user_history (model=LogDwellTimeBigrams, dwelltime more than 120sec)

Top2 weighted average cosine similarity between query embedding and clustered embedding of documents from user_history (model=LogDwellTimeBigrams, dwelltime less than 120sec)

cnt / (1 + cnt), cnt = sum of weights where cosine between query embedding and clustered embedding of documents from user_history > 0.30 (model=LogDwellTimeBigrams, dwelltime more than 120sec)

cnt / (1 + cnt), cnt = sum of weights where cosine between query embedding and clustered embedding of documents from user_history > 0.30 (model=LogDwellTimeBigrams, dwelltime less than 120sec)

cnt / (1 + cnt), cnt = sum of weights where cosine between query embedding and clustered embedding of documents from user_history > 0.60 (model=LogDwellTimeBigrams, dwelltime more than 120sec)

cnt / (1 + cnt), cnt = sum of weights where cosine between query embedding and clustered embedding of documents from user_history > 0.60 (model=LogDwellTimeBigrams, dwelltime less than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime less than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime less than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime more than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime less than 120sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime more than 120sec)

DotProducts = (all click embeddings from user_history) * QueryEmb, Take top1 weighted DotProduct

DotProducts = (all click embeddings from user_history) * DocEmb, Take top1 weighted DotProduct

DotProducts = (all click embeddings from user_history where dwelltime < 60) * DocEmb, Take top1 weighted DotProduct

DotProducts = (all click embeddings from user_history where dwelltime < 60) * DocEmb, Take DotProducts > 0.50, f = count / count + 10

DotProducts = (all click embeddings from user_history where dwelltime < 60) * DocEmb, min DotProducts

DotProducts = (all click embeddings from user_history where dwelltime >= 60) * DocEmb, Take average of top2 weighted DotProducts

DotProducts = (all click embeddings from user_history where dwelltime >= 60) * DocEmb, Take average of top4 weighted DotProducts

DotProducts = (all click embeddings from user_history where dwelltime >= 90) * DocEmb, Take average of top2 weighted DotProducts

DotProducts = (all click embeddings from user_history where dwelltime >= 90) * DocEmb, Take DotProducts > 0.40, f = count / count + 10

DotProducts = (all click embeddings from user_history where dwelltime >= 90) * DocEmb, sort descending, f = DotProducts[0.4 * len(DotProducts)]

DotProducts = (all click embeddings from user_history where dwelltime >= 180) * DocEmb, Take average of top4 weighted DotProducts

DotProducts = (all click embeddings from user_history where dwelltime >= 210) * DocEmb, sort descending, f = DotProducts[0.1 * len(DotProducts)]

DotProducts = (all click embeddings from user_history where dwelltime >= 450) * DocEmb, Take DotProducts > 0.10, f = 2 * count / (count + len(DotProducts))

DotProducts = (longer 120s click embeddings from user_history where dwelltime >= 600) * QueryEmb, sort descending, f = DotProducts[0.2 * len(DotProducts)]

DotProducts = (longer 120s click embeddings from user_history) * DocEmb, max DotProducts

DotProducts = (longer 120s click embeddings from user_history where dwelltime < 180) * DocEmb, Take average of top(0.30 * len(DotProducts)) weighted DotProducts

DotProducts = (longer 120s click embeddings from user_history where dwelltime < 180) * DocEmb, sort descending, f = DotProducts[0.9 * len(DotProducts)]

DotProducts = (longer 120s click embeddings from user_history where dwelltime >= 180) * DocEmb, Take top1 weighted DotProduct

DotProducts = (longer 120s click embeddings from user_history where dwelltime >= 180) * DocEmb, Take average of top2 weighted DotProducts

DotProducts = (longer 120s click embeddings from user_history where dwelltime >= 360) * DocEmb, Take top1 weighted DotProduct

DotProducts = (longer 120s click embeddings from user_history where dwelltime >= 360) * DocEmb, Take average of top(0.10 * len(DotProducts)) weighted DotProducts

DotProducts = (longer 120s click embeddings from user_history where dwelltime >= 360) * DocEmb, Take average of top2 weighted DotProducts

DotProducts = (longer 120s click embeddings from user_history where dwelltime >= 360) * DocEmb, sort descending, f = DotProducts[0.2 * len(DotProducts)]

DotProducts = (longer 120s click embeddings from user_history where dwelltime < 600) * DocEmb, Take top1 weighted DotProduct

DotProducts = (longer 120s click embeddings from user_history where dwelltime < 600) * DocEmb, Take average of top2 weighted DotProducts

DotProducts = (longer 120s click embeddings from user_history where dwelltime < 600) * DocEmb, Take DotProducts > 0.50, f = count / count + 10

DotProducts = (longer 120s click embeddings from user_history where dwelltime >= 600) * DocEmb, Take average of top2 weighted DotProducts

DotProducts = (longer 120s click embeddings from user_history where dwelltime >= 600) * DocEmb, Take DotProducts > 0.40, f = weightSum / (weightSum + 1)

DotProducts = (longer 120s click embeddings from user_history where dwelltime >= 600) * DocEmb, Take DotProducts > 0.10, f = 2 * count / (count + len(DotProducts))

DotProducts = (longer 120s click embeddings from user_history where dwelltime >= 600) * DocEmb, max DotProducts

Cosine similarity between query embedding and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime less than 30sec)

Cosine similarity between query embedding and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 30sec)

Cosine similarity between query embedding and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime less than 300sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 30sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 300sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 600sec)

Cosine similarity between document and fading embedding of documents from RTMR user_history (model=LogDwellTimeBigrams, fadingCoef=12days, dwelltime more than 90sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 30sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime less than 90sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 300sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.2days, dwelltime less than 600sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.3days, dwelltime more than 30sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.3days, dwelltime more than 600sec)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=300days, dwelltime less than 120sec DWT30)

Take clustered embeddings (short clicks, centroids with weights less than 5), Calc dot products with query embedding, f = Min(DotProducts)

Take clustered embeddings (short clicks), Calc dot products with doc embedding, Take dot product more than 0.95, f = 1 / (1 + sum of weights)

Take clustered embeddings (long clicks, centroids with weights less than 5), Calc dot products with query embedding, sort descending, f = DotProducts[0.10 * DotProduct.size()]

Take clustered embeddings (long clicks, centroids with weights less than 5), Calc dot products with query embedding, sort descending, f = DotProducts[0.90 * DotProduct.size()]

Take clustered embeddings (long clicks, centroids with weights less than 5), Calc dot products with query embedding, f = Min(DotProducts)

Take clustered embeddings (long clicks, centroids with weights less than 5), Calc dot products with query embedding, Take dot product more than 0.30, f = 1 / (1 + sum of weights)

Take clustered embeddings (long clicks, centroids with weights less than 5), Calc dot products with query embedding, Take dot product more than 0.40, f = count / (count + 10)

Take clustered embeddings (long clicks, centroids with weights less than 5), Calc dot products with query embedding, Take dot product more than 0.40, f = 1 / (1 + sum of weights)

Take clustered embeddings (long clicks, centroids with weights less than 10), Calc dot products with query embedding, sort descending, take top5, f = AVG(DotProduct[i] * Weight[i])

Take clustered embeddings (long clicks), Calc dot products with doc embedding, f = Max(DotProduct)

Take clustered embeddings (long clicks), Calc dot products with doc embedding, f = AVG(DotProduct[i] * Weight[i])

Take clustered embeddings (long clicks), Calc dot products with doc embedding, Take dot product more than 0.95, f = 1 / (1 + sum of weights)

Take clustered embeddings (long clicks, centroids with weights less than 5), Calc dot products with doc embedding, f = Max(DotProduct)

Take clustered embeddings (long clicks, centroids with weights less than 5), Calc dot products with doc embedding, sort descending, f = DotProducts[0.05 * DotProduct.size()]

Take clustered embeddings (long clicks, centroids with weights less than 5), Calc dot products with doc embedding, sort descending, f = DotProducts[0.30 * DotProduct.size()]

Take clustered embeddings (long clicks, centroids with weights less than 5), Calc dot products with doc embedding, f = Min(DotProducts)

Take clustered embeddings (long clicks, centroids with weights less than 5), Calc dot products with doc embedding, sort descending, take top5, f = AVG(DotProduct[i] * Weight[i])

Take clustered embeddings (long clicks, centroids with weights less than 5), Calc dot products with doc embedding, Take dot product more than 0.30, f = 1 / (1 + sum of weights)

Take clustered embeddings (long clicks, centroids with weights more than 5), Calc dot products with doc embedding, sort descending, f = DotProducts[0.10 * DotProduct.size()]

Take clustered embeddings (long clicks, centroids with weights less than 10), Calc dot products with doc embedding, f = Max(DotProduct)

Take clustered embeddings (long clicks, centroids with weights less than 10), Calc dot products with doc embedding, sort descending, f = DotProducts[0.10 * DotProduct.size()]

Take clustered embeddings (long clicks, centroids with weights less than 20), Calc dot products with doc embedding, f = AVG(DotProduct[i] * Weight[i])

Take clustered embeddings (long clicks, centroids with weights less than 20), Calc dot products with doc embedding, Take dot product more than 0.70, f = 1 / (1 + sum of weights)

Take clustered embeddings (long clicks, centroids with weights less than 30), Calc dot products with doc embedding, f = Max(DotProduct)

Take clustered embeddings (long clicks, centroids with weights less than 30), Calc dot products with doc embedding, sort descending, take top20, f = AVG(DotProduct[i] * Weight[i])

Cosine similarity between query embedding and fading embedding of queries from RTMR user_history, (model=LogDwellTimeBigrams, fadingCoef=0.01days)

Cosine similarity between query embedding and fading embedding of queries from RTMR user_history, (model=LogDwellTimeBigrams, fadingCoef=0.02days)

Cosine similarity between doc embedding and fading embedding of queries from RTMR user_history, (model=LogDwellTimeBigrams, fadingCoef=0.01days)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.03days)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=1.2days)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=6days)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=14days)

Take clustered embeddings (query embeddings, all weights), Calc dot product with query embedding, Take top4, f = AVG(Score[i] * Weight[i])

Take clustered embeddings (query embeddings, all weights), Calc dot product with query embedding, Take dot product with score > 0.75, f = 1 / (1 + sum of weights)

Take clustered embeddings (query embeddings, weights less than 3), Calc dot product with query embedding, Sort descending, f = DotProduct[0.95 * length(DotProducts)]

Take clustered embeddings (query embeddings, weights less than 5), Calc dot product with query embedding, Take dot product with score > 0.55, f = 1 / (1 + sum of weights)

Take clustered embeddings (query embeddings, weights less than 5), Calc dot product with query embedding, f = MIN(DotProducts)

Take clustered embeddings (query embeddings, weights less than 10), Calc dot product with query embedding, f = MAX(DotProducts)

Take clustered embeddings (query embeddings, weights less than 15), Calc dot product with query embedding, f = MAX(Score[i] * Weights[i])

Take clustered embeddings (query embeddings, weights less than 20), Calc dot product with query embedding, Take dot product with score > 0.70, f = 1 / (1 + sum of weights)

Take clustered embeddings (query embeddings, all weights), Calc dot product with query embedding, f = MAX(Score[i] * Weights[i])

Take clustered embeddings (query embeddings, weights less than 3), Calc dot product with doc embedding, Sort descending, f = DotProduct[0.05 * length(DotProducts)]

Take clustered embeddings (query embeddings, weights less than 3), Calc dot product with doc embedding, f = MIN(DotProducts)

Take clustered embeddings (query embeddings, weights less than 5), Calc dot product with doc embedding, f = MIN(DotProducts)

Take clustered embeddings (query embeddings, weights less than 15), Calc dot product with doc embedding, Take dot product with score > 0.45, f = 1 / (1 + sum of weights)

Take clustered embeddings (query embeddings, all weights), Calc dot product with doc embedding, f = MAX(Score[i] * Weights[i])

Take clustered embeddings (query embeddings, all weights), Calc dot product with doc embedding, Take dot product with score > 0.55, f = 1 / (1 + sum of weights)

Take last 10 query embeddings, Calc dot product with query embedding, f = MAX(Score[i] * Weights[i])

Take last 10 query embeddings, Calc dot product with doc embedding, f = MAX(Score[i] * Weights[i])

Take last 20 query embeddings, Calc dot product with query embedding, f = MAX(Score[i] * Weights[i])

Take last 20 query embeddings, Calc dot product with query embedding, Take top3, f = AVG(Score[i] * Weight[i])

Take last 30 query embeddings, Calc dot product with query embedding, Sort descending, f = DotProduct[0.05 * length(DotProducts)]

Take last 40 query embeddings, Calc dot product with query embedding, f = MAX(Score[i] * Weights[i])

Take last 50 query embeddings, Calc dot product with query embedding, f = MAX(DotProducts)

Take last 50 query embeddings, Calc dot product with query embedding, sort descending, f = DotProducts[1] (in zero-numeration)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime less than 120sec), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 120sec), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime less than 120sec), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.5days, dwelltime more than 120sec), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime less than 120sec), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=3days, dwelltime more than 120sec), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 30sec), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime less than 90sec), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.1days, dwelltime more than 300sec), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.2days, dwelltime less than 600sec), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.3days, dwelltime more than 30sec), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.3days, dwelltime more than 600sec), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=300days, dwelltime less than 120sec DWT30), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=0.03days), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=1.2days), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=6days), f = 1 - 2^(-0.2 * value)

fadingEmbedding.Norm * CalcFading(fadingEmbedding.FadingCoef, requestTimestamp - fadingEmbedding.LastUpdTimestamp), (model=LogDwellTimeBigrams, fadingCoef=14days), f = 1 - 2^(-0.2 * value)

Position (1- Fieldset3bclmweightedflogw0k0001) * (1 - DSSMLONGMIDLESHORDSHARDCLICS) Among all documents on PRS

Max metafactor on Production:DssmMiddleVsShortLongHardNoClicks(1221)

Max metafactor on Production:DssmLogDwellTimeBigrams(1338)

DFT_GREATER_CNT metafactor on Production:DssmLogDwellTimeBigrams(1338)

DFT_SUM_WF_NORM_SUM_W metafactor on Production:DssmLogDwellTimeBigrams(1338)

Similar documents, type of similarity: mutualserp. SUMWFNORMSUMW AGGENT DSSMBIGRAMSQUERYDERIVATIMAX factor

Similar documents, type of similarity: mutualserp. Max Aggregion DSSMLOGDWELTIMEBIGRAMS Factor

Similar documents, type of similarity: mutualserp. Sumwfnormsumw DSSMLONGVSMIDLESHORTNOCLICS AGGENT

Similar documents, type of similarity: mutualserp. Max Agnigation DSSMLONGMIDLESHORDSHARDCLICS Factor

Max metafactor on Production:DssmBoostingXfWeightKMeans5AvgTop02ScoreQE(1481)

DFT_MAX metafactor on Production:DssmBoostingXfWeightKMeans5AvgTop02ScoreQE(1481)

DFT_GREATER_CNT metafactor on Production:DssmBoostingXfWeightKMeans5AvgTop02ScoreQE(1481)

Fract metafactor on Production:DssmBoostingXfOneSeKMeans1Score(1489)

Meta:Max metafactor on web_production:DssmBoostingXfOneSeAmSsHardKMeans1Score(1597)

Meta:Fract metafactor on web_production:DssmBoostingXfOneSeAmSsHardKMeans1Score(1597)

SD:DFT_MAX metafactor on web_production:DssmBoostingXfOneSeAmSsHardKMeans1ScoreAvgClusterTop3Weighted(1598)

Model trained on url, title and user regions chain. Target: DwellTime

The proximity between the title and the word 'reviews', calculated using the model from the Factor-1635

The proximity between the title and the word 'News', designed using the model from the Factor-1635

The proximity between the title and the word 'how much', calculated using the model from the Factor-1635

The proximity between the title and the word 'why', calculated using the model from the Factor-1635

The proximity between the title and the word 'do', designed using the model from the Factor-1635

The proximity between the title and the word 'Instagram', designed using the model from the Factor-1635

The proximity between the title and the word 'Ukraine', calculated using the model from the Factor-1635

The proximity between the title and the word 'meaning', calculated using the model from the Factor-1635

The proximity between the title and the word 'Beautiful', calculated using the model from the Factor-1635

The proximity between the title and the word 'stands', calculated using the model from the Factor-1635

The minimum proximity between the title and the words of the request calculated using the model from the Factor-1635

The difference in maximum and minimal proximity between the title and all the words of the request

We sort the words of the proximity to Title and leave half the closest. The meaning of the factor is the average proximity among the remaining (closest to Title)

Similarly, fi_query_word_title_Query_min_word_similarity (minimum proximity between the title and the words of the request), but the numbers in the request are not taken into account

The maximum proximity between the title and the words of the city of the user, calculated using the model from the Factor-1635

The minimum proximity between the title and the words of the user area calculated using the model from the Factor-1635

The maximum proximity between the title and the words of the user area calculated using the model from the Factor-1635

A request factor calculated using a model from the Factor-1635. The minimum similarity between all words of the request.

The proximity between the title and the word 'series', calculated using the model from the Factor-1635

The proximity between the title and the word 'Official', calculated using the model from the Factor-1635

The proximity between the title and the word 'season', calculated using the model from the Factor-1635

The proximity between the title and the word 'birth', calculated using the model from the Factor-1635

The proximity between the title and the word 'good', calculated using the model from the Factor-1635

Meta:Avg metafactor on web_production:DssmQueryDwellTime(1406)

Meta:Rmse metafactor on web_production:DssmLogDtBigramsAMHardQueriesNoClicks(1523)

Meta:ResidMax metafactor on web_production:XfDtShowKnnTopSumW2FSumWFieldSet1Bm15FLogK0001(1583)

Meta:ResidMax metafactor on web_production:DssmLogDtBigramsAMHardQueriesNoClicksMixed(1596)

Meta:EpsHashShare metafactor on web_production:DssmLogDtBigramsAMHardQueriesNoClicksMixed(1596)

SD:DFT_MAX metafactor on web_production:DssmBoostingXfOneSeAmSsHardKMeans1Score(1597)

Meta:ResidMax metafactor on web_production:DssmBoostingXfOneSeAmSsHardKMeans1Score(1597)

Meta:Resid metafactor on web_production:DssmQueryDwellTime(1406)

Meta:Resid metafactor on web_production:DssmLogDtBigramsAMHardQueriesNoClicksMixed(1596)

The proximity between the title and the word 'vk', calculated using the model from the Factor-1635

The proximity between the title and the word 'buy', designed using the model from the Factor-1635

The proximity between the title and the word 'music', designed using the model from the Factor-1635

Fract metafactor on NeuroTextModelLongClickPredictorByWordAndBigramCountersWithSSHards(1845).

Resid metafactor on NeuroTextModelLongClickPredictorByWordAndBigramCountersWithSSHards(1845).

EpsHashShare metafactor on NeuroTextModelLongClickPredictorByWordAndBigramCountersWithSSHards(1845).

Meta:Fract metafactor on web_production:DssmBoostingSerpSimilarityHardKMeans1Score(1841)

Meta:Resid metafactor on web_production:DssmCtrEngSsHard(1855)

Similar documents, type of similarity: mutualserp. Greatercnt ReformulationslonGestClicklogdt Agnigation Factor

Similar documents, type of similarity: mutualserp. SUMWFNORMSUMW ReformulationslongestClicklogdt factor

bert model which predicts 1rel 2vital 0other for arabic alice music search scenario

slot for support of two parallel aramusic models

small-bert-model sbr:1721013142 predict Predict_sinsig_mse_1502766203_standartized_mse_target - Distillation of the sinsig mse bert model 1502766203

small-bert-model sbr:1721013142 predict Predict_dbd_mse_1502778723_standartized_mse_target Distillation of the dbd mse bert model 1502778723

small-bert-model sbr:1721013142 predict Predict_sinsig_fresh_20200118_0502_1528253243_standartized_mse_target Distillation of the fresh bert model 1528253243

small-bert-model sbr:1721013142 predict Predict_sinsig_ce_multitarget_012_1511368337_standartized_mse_target Distillation of the sinsig ce multitarget bert model 1511368337, head 0.12 threshold

small-bert-model sbr:1721013142 predict Predict_sinsig_ce_multitarget_025_1511368337_standartized_mse_target Distillation of the sinsig ce multitarget bert model 1511368337, head 0.25 threshold

small-bert-model sbr:1721013142 predict Predict_sinsig_ce_multitarget_05_1511368337_standartized_mse_target Distillation of the sinsig ce multitarget bert model 1511368337, head 0.5 threshold

small-bert-model sbr:1721013142 predict Predict_proxima_large_mse_1545302456_standartized_mse_target Distillation of the proxima mse bert model 1545302456

small-bert-model sbr:1721013142 predict Predict_click_pers_ce_1534051635_standartized_mse_target Distillation of the click pers ce bert model 1534051635

small-bert-model sbr:1721013142 predict Predict_click_odd_mse_1512795366_standartized_mse_target Distillation of the click odd mse bert model 1512795366

split-bert-model sbr:2005741938 predict Predict_mse_prediction_1628966206_proxima_standartized Split distillation of the proxima mse bert model 1628966206

split-bert-model sbr:2005741938 predict Predict_ce_0_12_1633847052_sinsig_standartized Split distillation of the sinsig ce multitarget bert model 1633847052, head 0.12 threshold

split-bert-model sbr:2005741938 predict Predict_mse_Predict_ce_0_25_1633847052_sinsig_standartized Split distillation of the sinsig ce multitarget bert model 1633847052, head 0.25 threshold

split-bert-model sbr:2005741938 predict Predict_mse_Predict_ce_0_5_1633847052_sinsig_standartized Split distillation of the sinsig ce multitarget bert model 1633847052, head 0.5 threshold

split-bert-model sbr:2005741938 predict mse_Predict_ce_0_78_1633847052_sinsig_standartized Split distillation of the sinsig ce multitarget bert model 1633847052, head 0.78 threshold

split-bert-model sbr:2005741938 predict Predict_mse_Predict_ce_1_3_1633847052_sinsig_standartized Split distillation of the sinsig ce multitarget bert model 1633847052, head 1.3 threshold

split-bert-model sbr:2005741938 predict Predict_mse_Predict_ce_2_0_1633847052_sinsig_standartized Split distillation of the sinsig ce multitarget bert model 1633847052, head 2.0 threshold

split-bert-model sbr:2005741938 predict Predict_mse_prediction_1625097377_proxima_standartized Split distillation of the proxima mse bert model 1625097377

split-bert-model sbr:2005741938 predict Predict_mse_prediction_1633307071_dbd_standartized Split distillation of the dbd mse bert model 1633307071

split-bert-model sbr:2005741938 predict prediction_1528253243_sinsig_fresh_20200118_0502_standartized Split distillation of the fresh bert model 1528253243

split-bert-model sbr:2005741938 predict prediction_1644380449_is_pirate_0807_standartized Split distillation of the is_pirate bert model 1644380449

split-bert-model sbr:2005741938 predict Predict_oct_large_1871529228_sinsig_relev10_mse_stand_mse Split distillation of the nov-large model with relev10 1871529228

split-bert-model sbr:2005741938 predict Predict_oct_large_1872243553_proxima_relev10_mse_stand_mse Split distillation of the nov-large model with relev10 1872243553

split-bert-model sbr:2005741938 predict Predict_xLarge_prediction_nov_lav_1990027087_standartized_mse Split distillation of the xlarge model with relev10 1990027087

split-bert-model sbr:2005741938 predict Predict_xLarge_prediction_nov_proxima_1970068024_standartized_mse Split distillation of the nov-large model with relev10 1970068024

split-bert-model sbr:2005741938 predict Predict_Large_predict_target_fin_law_1967416198_standartized_mse Split distillation of the large model 1967416198

split-bert-model sbr:2005741938 predict Predict_Large_predict_target_med_doc_1967182618_standartized_mse Split distillation of the large model 1967182618

split-bert-model sbr:2005741938 predict Predict_Large_predict_target_sos_1969262884_standartized_mse Split distillation of the large model 1969262884

sbr:2411087272 splitv3 head Predict_jul_xlarge_lav_with_platform_2344793612_stand_mse

sbr:2411087272 splitv3 head Predict_jul_xlarge_sinsig_baskets_with_platform_2339818891_stand_mse

berts_storage:2022/02/artmironov/2831823537 splitv4 head Predict_bert_650_merge_platform_no_trans_v1_2784077050_stand_mse

bert model 2021/09/boyalex/2405042847 head: Predict_jul_xlarge_sinsig_baskets_2339818891_stand_mse

bert model 2021/09/boyalex/2405042847 head: Predict_large_cs_clever_clicks_2360931838_stand_mse

bert model 2021/09/boyalex/2405042847 head: Predict_large_google_2334139725_stand_mse

bert model 2021/09/boyalex/2405042847 head: Predict_pq_large_2394064809_stand_mse

bert model 2021/09/boyalex/2405042847 head: Predict_jul_xlarge_cs_sinsig_full_2394990987_stand_mse

bert model 2021/09/boyalex/2405042847 head: Predict_jul_xlarge_cs_apscore_2397363137_stand_mse

The probability that the VK.com host is popular for this request in accordance with the corresponding DSSM model.

The probability that the Onliner.by host is popular for this request according to the corresponding DSSM model.

The probability that the Rambler.ru host is popular for this request in accordance with the corresponding DSSM model.

The likelihood that the ExpertCen.ru host is popular for this request in accordance with the corresponding DSSM model.

The probability that the Sunhome.ru host is popular for this request according to the corresponding DSSM model.

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

The main components of the requesting Embling from the DSSMCTRNOMINER model

Predicted by demand and country, using a DSSM model, the length of the click from this country.

The average for the year for the year predicted using the neural network.

ADDTIME ADDTIME is predicted using a neural network for a year.

The average Txthirelesy value predicted using a neural network for the year.

The average Textlike is predicted using a neural network for the year.

The average HasnoallwordStrsy is predicted using a neural network for a year.

The average ISFORUM is predicted using a neural network for the year.

The average Haspayments is predicted using a neural network for the year.

The average value of Yabarhostavgtime2 for the year for the year.

The average yabarurlvisitors is predicted using a neural network for the year.

The average value of QueryDowneronlyClickRate for the year for the year.

The average Dateraage value for the year for a year predicted using a neural network.

The average LonGestText is predicted using a neural network for the year.

The average DifferentinTernallinks for the year for the year.

The average value of QueryDowneronlyClickRate_Rreg is predicted using a neural network for a year.

The average ISTARGETBUSSINESSCARD is predicted using a neural network for the year.

The average BOCM value predicted using a neural network for the year.

The average ISindEXPAGE is predicted using a neural network for the year.

The average value of QueriesavGCM2 for the year for the year predicted using a neural network.

The average BrowserhostdowLoadProbabolyti is predicted using a neural network for the year.

The average value of Regbrowseruserhub for the year for a year predicted using a neural network.

The average AuxtitlebM25 average value for the year for the year.

The average value of QueryurlCorrectrxFactor for the year for the year.

The average value of QueryTodoCallsumfcountTextbM11Norm16384 for the year for the year.

The average value of the XFDTSHOWALSUMWFSUMWBODYMINWINDOWSIZE for the year for the year.

The value of the ISMAINPAGE with clicks predicted using the neural network with clicks on request for the year.

A mid Yabarurlavgtime value predicted using a neural network with clicks for a year.

DiffferentinTernallinks, which is predicted using a neural network, is a weighted net with clicks for a year.

The Malue Network DwellTime-AMI predicted using the neural network is the value of Urldomainfraction for the year.

The prediction of the probability that the request is localized in accordance with the regionality5 rule.

Characterizes the request for the degree of change from removing a fixed word ('site' for Kirilitsa), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

Characterizes the request for the degree of change from the addition of a fixed word (number of some year), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

Characterizes a request for the degree of change from the addition of a fixed word ('online' for Kirilitsa), DSSM model DSSMBOOSTINGXFONESEAMSARD is used

DSSM prediction of google specificity for query

The average value of Randomlogqueryavgaddtime of the closest KNN queries.

The average value of RandomlogqueryavgtXthirelsy nearest KNN queries.

The average value of Randomlogqueryavgtextlike nearest KNN queries.

The average value of Randomlogqueryavgisforum of the closest KNN queries.

The average value of Randomlogqueryavghaspayments closest to KNN queries.

The average value of Randomlogqueryavgdiferentinternallinks of the nearest KNN queries.

The average value of RandomlogqueryavgistargetbussinessCard of the nearest KNN queries.

The average value is RandomlogqueryavgquerytododoCallsumfcountTextbM11NORM16384 of the nearest KNN queries.

The average value is Randomlogqueryavgxfdtshowallsumwfsumwbodyminwindowsize closest KNN queries.

The average value of the request factor according to lingvobusting - Querytotextbyxfdtshowknn, is calculated in the Hippo rule LingBoostqueryFeatures

Quantile 0.1 for the request factor according to the XFDTSHOWKNN linguosting data, is calculated in the LingBoostqueryFeatures Hotemaway Rules

Quantile 0.9 for the quantity factor according to the XFDTSHOWKNN linguosting data, is calculated in the LingBoostqueryFeatures Hotemaway Rules

Newest runtime factor nn_over_dssm_dot_product. L0-embed relevance value, b-embed.

A model trained to predict an assessment of the USSR-DUMP-20191206 PRS-20191207 ALL-4-YEARS QMSE 20K 0.2 -s 0.8 -z 1 formula.

A model trained to predict an assessment of the USSR-DUMP-20191206 PRS-20191207 ALL-8-YEARS [T> 0.25] Crossentropy 20K 0.25 -s 0.8 -z 1.

A model trained to predict an assessment of the USSR-DUMP-20191206 PRS-20191207 Yang-Relevance-2-Years [T> 0.5] Crossentropy 20K 0.25 -s 0.8 -z 1.

ODD target DSSM on top 200 catboost fstr features

Log DT target DSSM on top 200 catboost fstr features