公开(公告)号：US11195023B2

公开(公告)日：2021-12-07

申请号：US16024814

申请日：2018-06-30

Applicant: Microsoft Technology Licensing, LLC

Inventor： Christopher Wright Lloyd, II ,  Konstantin Salomatin ,  Jeffrey Douglas Gee ,  Mahesh S. Joshi ,  Shivani Rao ,  Vladislav Tcheprasov ,  Gungor Polatkan ,  Deepak Kumar Dileep Kumar

IPC: G06K9/00 ,  G06N20/00 ,  G06K9/62

Abstract: Techniques for implementing a feature generation pipeline for machine learning are provided. In one technique, multiple jobs are executed, each of which computes a different set of feature values for a different feature of multiple features associated with videos. A feature registry is stored that lists each of the multiple features. After the jobs are executed and the feature registry is stored, a model specification is received that indicates a set of features for a model. For each feature in a subset of the set of features, a location is identified in storage where a value for said each feature is found and the value for that feature is retrieved from the location. A feature vector is created that comprises, for each feature in the set of features, the value that corresponds to that feature. The feature vector is used to train the model or as input to the model.

公开(公告)号：US10602226B2

公开(公告)日：2020-03-24

申请号：US16020843

申请日：2018-06-27

Applicant: Microsoft Technology Licensing, LLC

Inventor： Gungor Polatkan ,  Yulia Astakhova ,  Deepak Kumar ,  Konstantin Salomatin ,  Jeffrey Douglas Gee ,  Mahesh S. Joshi ,  Shivani Rao

IPC: H04N21/47 ,  H04N21/466 ,  H04N21/478 ,  H04N21/45 ,  H04N21/475 ,  H04N21/237 ,  H04N21/24 ,  H04N21/2743 ,  H04N21/25 ,  H04N21/258 ,  G06Q50/00

Abstract: The recommendation system provided with an on-line connection system identifies on-line recommendations of videos and generates a user interface (UI) by including into the resulting presentation selected recommendations of videos. The recommendations of videos presented in the UI are organized into groups that are topically coherent, where each group is decorated with a context annotation—an explanation of why the recommendations in a given carousel are relevant for a member. Each video that is being evaluated by the recommendation system with respect to a subject member profile is assigned an annotation that is selected from a plurality of potentially applicable annotations. The technical problem of optimizing an order of presentation of recommendations grouped by context annotations, in a UI where annotations drive the layout, is addressed by deriving ranks for different context annotations based on global and personalized click through rates and using these values assigned to respective context annotations in constructing the UI.

公开(公告)号：US20200007937A1

公开(公告)日：2020-01-02

申请号：US16020843

申请日：2018-06-27

Applicant: Microsoft Technology Licensing, LLC

Inventor： Gungor Polatkan ,  Yulia Astakhova ,  Deepak Kumar ,  Konstantin Salomatin ,  Jeffrey Douglas Gee ,  Mahesh S. Joshi ,  Shivani Rao

IPC: H04N21/475 ,  H04N21/466 ,  H04N21/478

Abstract: The recommendation system provided with an on-line connection system identifies on-line recommendations of videos and generates a user interface (UI) by including into the resulting presentation selected recommendations of videos. The recommendations of videos presented in the UI are organized into groups that are topically coherent, where each group is decorated with a context annotation—an explanation of why the recommendations in a given carousel are relevant for a member. Each video that is being evaluated by the recommendation system with respect to a subject member profile is assigned an annotation that is selected from a plurality of potentially applicable annotations. The technical problem of optimizing an order of presentation of recommendations grouped by context annotations, in a UI where annotations drive the layout, is addressed by deriving ranks for different context annotations based on global and personalized click through rates and using these values assigned to respective context annotations in constructing the UI.

公开(公告)号：US12223554B2

公开(公告)日：2025-02-11

申请号：US16912245

申请日：2020-06-25

Applicant: Microsoft Technology Licensing, LLC

Inventor： Rohan Ramanath ,  Konstantin Salomatin ,  Jeffrey Douglas Gee ,  Onkar Anant Dalal ,  Gungor Polatkan ,  Sara Smoot Gerrard ,  Deepak Kumar ,  Rupesh Gupta ,  Jiaqi Ge ,  Lingjie Weng ,  Shipeng Yu

IPC: G06Q50/00 ,  G06F16/9535 ,  G06F16/958 ,  G06F18/214 ,  G06N5/04 ,  G06Q10/1053

Abstract: In some embodiments, a computer system generates a recommendation for a user of an online service based on user actions that have been performed by the user within a threshold amount of time before the generation of the recommendation. For each user action, the computer system determines an intent classification that identifies an activity of the user and that corresponds to different types of user actions, as well as a preference classification that identifies a target of the activity, and then stores these intent and preference classifications as part of indications of the user actions for use in generating different types of recommendations using different types of recommendation models. Additionally, the computer system may use mini-batches of data from an incoming stream of logged data to train an incremental update to one or more recommendation models.

公开(公告)号：US10887655B2

公开(公告)日：2021-01-05

申请号：US16020260

申请日：2018-06-27

Applicant: Microsoft Technology Licensing, LLC

Inventor： Konstantin Salomatin ,  Fares Hedayati ,  Jeffrey Douglas Gee ,  Mahesh S. Joshi ,  Shivani Rao ,  Gungor Polatkan ,  Deepak Kumar

IPC: H04N21/475 ,  G06K9/62 ,  H04N21/482 ,  H04N21/466 ,  H04N21/254 ,  H04N21/25 ,  H04N21/47 ,  H04N21/2543 ,  H04N21/258

Abstract: The video recommendation system provided with an on-line connection system generates on-line video recommendations using collaborative filtering for clusters of member profiles. The recommendation system clusters member profiles using member profile information as clustering criteria. The video recommendations are then generated for a given cluster, based on aggregation of video viewing history recorded for the member profiles that are in the given cluster, using the video similarity matrix. In order to produce video recommendations for a particular member profile, the recommendation system first determines cluster membership for the member profile, retrieves recommendations generated for that cluster, and provides recommendations to the associated member. A user interface including references to one or more recommended videos is rendered on a display device of a viewer.

公开(公告)号：US20200005045A1

公开(公告)日：2020-01-02

申请号：US16024814

申请日：2018-06-30

Applicant: Microsoft Technology Licensing, LLC

Inventor： Christopher Wright Lloyd, II ,  Konstantin Salomatin ,  Jeffrey Douglas Gee ,  Mahesh S. Joshi ,  Shivani Rao ,  Vladislav Tcheprasov ,  Gungor Polatkan ,  Deepak Kumar Dileep Kumar

IPC: G06K9/00 ,  G06K9/62 ,  G06F15/18

Abstract: Techniques for implementing a feature generation pipeline for machine learning are provided. In one technique, multiple jobs are executed, each of which computes a different set of feature values for a different feature of multiple features associated with videos. A feature registry is stored that lists each of the multiple features. After the jobs are executed and the feature registry is stored, a model specification is received that indicates a set of features for a model. For each feature in a subset of the set of features, a location is identified in storage where a value for said each feature is found and the value for that feature is retrieved from the location. A feature vector is created that comprises, for each feature in the set of features, the value that corresponds to that feature. The feature vector is used to train the model or as input to the model.

公开(公告)号：US20240112281A1

公开(公告)日：2024-04-04

申请号：US17952095

申请日：2022-09-23

Applicant: Microsoft Technology Licensing, LLC

Inventor： Keqing Liang ,  Konstantin Salomatin ,  Noureddine El Karoui

IPC: G06Q50/00 ,  G06Q30/02

CPC classification number: G06Q50/01 ,  G06Q30/0201

Abstract: In an example embodiment, a blending model is presented based on a linear programming approach. The blending model produces a slate of sponsored and non-sponsored pieces of content for display in a graphical user interface, with the ordering and placement of the sponsored and non-sponsored pieces of content selected in order to maximize an objective function. Such an approach can fine tune each piece of content using content-level parameters and holistically examine global constraints and opportunities. It establishes a robust optimization framework that can adapt to content and domain changes without requiring tuning through online experiments.

公开(公告)号：US20220245512A1

公开(公告)日：2022-08-04

申请号：US17167549

申请日：2021-02-04

Applicant: Microsoft Technology Licensing, LLC

Inventor： Kirill Talanine ,  Konstantin Salomatin ,  Arjun K. Kulothungun ,  Huseyin Baris Ozmen ,  Linda Fayad ,  Gungor Polatkan ,  Deepak Kumar Dileep Kumar

IPC: G06N20/00 ,  G06Q10/10 ,  G06F16/9535

Abstract: In an example embodiment, a fully automated process is provided for frequent model retraining and redeployment of a machine learned model trained to output a prediction of how likely it is that a candidate is qualified for a particular job posting. Model quality verification is provided by maintaining a snapshot of a baseline model and automatically comparing it to a proposed model by performing various metrics on the models by testing the models using a holdout data set that includes only data that was not used during the training process. Overlap between data in the holdout set used during retraining and the training set used during initial training is prevented by splitting each dataset using a hash on certain fields of the data.

公开(公告)号：US20200007936A1

公开(公告)日：2020-01-02

申请号：US16020260

申请日：2018-06-27

Applicant: Microsoft Technology Licensing, LLC

Inventor： Konstantin Salomatin ,  Fares Hedayati ,  Jeffrey Douglas Gee ,  Mahesh S. Joshi ,  Shivani Rao ,  Gungor Polatkan ,  Deepak Kumar

IPC: H04N21/475 ,  H04N21/466 ,  H04N21/482 ,  G06K9/62

Abstract: The video recommendation system provided with an on-line connection system generates on-line video recommendations using collaborative filtering for clusters of member profiles. The recommendation system clusters member profiles using member profile information as clustering criteria. The video recommendations are then generated for a given cluster, based on aggregation of video viewing history recorded for the member profiles that are in the given cluster, using the video similarity matrix. In order to produce video recommendations for a particular member profile, the recommendation system first determines cluster membership for the member profile, retrieves recommendations generated for that cluster, and provides recommendations to the associated member. A user interface including references to one or more recommended videos is rendered on a display device of a viewer.

公开(公告)号：US20210406838A1

公开(公告)日：2021-12-30

申请号：US16912245

申请日：2020-06-25

Applicant: Microsoft Technology Licensing, LLC

Inventor： Rohan Ramanath ,  Konstantin Salomatin ,  Jeffrey Douglas Gee ,  Onkar Anant Dalal ,  Gungor Polatkan ,  Sara Smoot Gerrard ,  Deepak Kumar ,  Rupesh Gupta ,  Jiaqi Ge ,  Lingjie Weng ,  Shipeng Yu

IPC: G06Q10/10 ,  G06N5/04 ,  G06K9/62 ,  G06F16/958 ,  G06F16/9535 ,  G06Q50/00

Abstract: In some embodiments, a computer system generates a recommendation for a user of an online service based on user actions that have been performed by the user within a threshold amount of time before the generation of the recommendation. For each user action, the computer system determines an intent classification that identifies an activity of the user and that corresponds to different types of user actions, as well as a preference classification that identifies a target of the activity, and then stores these intent and preference classifications as part of indications of the user actions for use in generating different types of recommendations using different types of recommendation models. Additionally, the computer system may use mini-batches of data from an incoming stream of logged data to train an incremental update to one or more recommendation models.