Hierarchical Modeling and Shrinkage for User Session LengthPrediction in Media Streaming
An important metric of users' satisfaction and engagement within on-line streaming services is the user session length, i.e. the amount of time they spend on a service continuously without interruption. Being able to predict this value directly benefits the recommendation and ad pacing contexts...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Published: |
Association for Computer Machinery
2019
|
| Online Access: | http://hdl.handle.net/1721.1/120991 https://orcid.org/0000-0003-1384-9743 |
| _version_ | 1826205721198329856 |
|---|---|
| author | Zhu, Zhen Vahabi, Hossein Dedieu, Antoine Mazumder, Rahul |
| author2 | Sloan School of Management |
| author_facet | Sloan School of Management Zhu, Zhen Vahabi, Hossein Dedieu, Antoine Mazumder, Rahul |
| author_sort | Zhu, Zhen |
| collection | MIT |
| description | An important metric of users' satisfaction and engagement within on-line streaming services is the user session length, i.e. the amount of time they spend on a service continuously without interruption. Being able to predict this value directly benefits the recommendation and ad pacing contexts in music and video streaming services. Recent research has shown that predicting the exact amount of time spent is highly nontrivial due to many external factors for which a user can end a session, and the lack of predictive covariates. Most of the other related literature on duration based user engagement has focused on dwell time for websites, for search and display ads, mainly for post-click satisfaction prediction or ad ranking. In this work we present a novel framework inspired by hierarchical Bayesian modeling to predict, at the moment of login, the amount of time a user will spend in the streaming service. The time spent by a user on a platform depends upon user-specific latent variables which are learned via hierarchical shrinkage. Our framework enjoys theoretical guarantees and naturally incorporates flexible parametric/nonparametric models on the covariates, including models robust to outliers. Our proposal is found to outperform state-of-the-art estimators in terms of efficiency and predictive performance on real world public and private datasets. |
| first_indexed | 2024-09-23T13:17:53Z |
| format | Article |
| id | mit-1721.1/120991 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T13:17:53Z |
| publishDate | 2019 |
| publisher | Association for Computer Machinery |
| record_format | dspace |
| spelling | mit-1721.1/1209912022-09-28T13:15:03Z Hierarchical Modeling and Shrinkage for User Session LengthPrediction in Media Streaming Zhu, Zhen Vahabi, Hossein Dedieu, Antoine Mazumder, Rahul Sloan School of Management Dedieu, Antoine Mazumder, Rahul An important metric of users' satisfaction and engagement within on-line streaming services is the user session length, i.e. the amount of time they spend on a service continuously without interruption. Being able to predict this value directly benefits the recommendation and ad pacing contexts in music and video streaming services. Recent research has shown that predicting the exact amount of time spent is highly nontrivial due to many external factors for which a user can end a session, and the lack of predictive covariates. Most of the other related literature on duration based user engagement has focused on dwell time for websites, for search and display ads, mainly for post-click satisfaction prediction or ad ranking. In this work we present a novel framework inspired by hierarchical Bayesian modeling to predict, at the moment of login, the amount of time a user will spend in the streaming service. The time spent by a user on a platform depends upon user-specific latent variables which are learned via hierarchical shrinkage. Our framework enjoys theoretical guarantees and naturally incorporates flexible parametric/nonparametric models on the covariates, including models robust to outliers. Our proposal is found to outperform state-of-the-art estimators in terms of efficiency and predictive performance on real world public and private datasets. 2019-03-15T18:23:31Z 2019-03-15T18:23:31Z 2018-10 2019-02-25T21:10:13Z Article http://purl.org/eprint/type/ConferencePaper 978-1-4503-6014-2 http://hdl.handle.net/1721.1/120991 Dedieu, Antoine, Rahul Mazumder, Zhen Zhu, and Hossein Vahabi. “Hierarchical Modeling and Shrinkage for User Session LengthPrediction in Media Streaming.” Proceedings of the 27th ACM International Conference on Information and Knowledge Management - CIKM ’18, 22-26 October, 2018, Torino, Italy, ACM, 2018. https://orcid.org/0000-0003-1384-9743 http://dx.doi.org/10.1145/3269206.3271700 Proceedings of the 27th ACM International Conference on Information and Knowledge Management - CIKM '18 Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Association for Computer Machinery arXiv |
| spellingShingle | Zhu, Zhen Vahabi, Hossein Dedieu, Antoine Mazumder, Rahul Hierarchical Modeling and Shrinkage for User Session LengthPrediction in Media Streaming |
| title | Hierarchical Modeling and Shrinkage for User Session LengthPrediction in Media Streaming |
| title_full | Hierarchical Modeling and Shrinkage for User Session LengthPrediction in Media Streaming |
| title_fullStr | Hierarchical Modeling and Shrinkage for User Session LengthPrediction in Media Streaming |
| title_full_unstemmed | Hierarchical Modeling and Shrinkage for User Session LengthPrediction in Media Streaming |
| title_short | Hierarchical Modeling and Shrinkage for User Session LengthPrediction in Media Streaming |
| title_sort | hierarchical modeling and shrinkage for user session lengthprediction in media streaming |
| url | http://hdl.handle.net/1721.1/120991 https://orcid.org/0000-0003-1384-9743 |
| work_keys_str_mv | AT zhuzhen hierarchicalmodelingandshrinkageforusersessionlengthpredictioninmediastreaming AT vahabihossein hierarchicalmodelingandshrinkageforusersessionlengthpredictioninmediastreaming AT dedieuantoine hierarchicalmodelingandshrinkageforusersessionlengthpredictioninmediastreaming AT mazumderrahul hierarchicalmodelingandshrinkageforusersessionlengthpredictioninmediastreaming |