Improving Scalability of Personalized Recommendation Systems for Enterprise Knowledge Workers
Enterprise knowledge workers have been overwhelmed by the growing rate of incoming data in recent years. In this paper, we present a recommendation system with the goal of helping knowledge workers in discovering useful new content. In particular, our system builds personalized user models based on...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2016-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/7368090/ |
| _version_ | 1818853908356792320 |
|---|---|
| author | Chetan Verma Michael Hart Sandeep Bhatkar Aleatha Parker-Wood Sujit Dey |
| author_facet | Chetan Verma Michael Hart Sandeep Bhatkar Aleatha Parker-Wood Sujit Dey |
| author_sort | Chetan Verma |
| collection | DOAJ |
| description | Enterprise knowledge workers have been overwhelmed by the growing rate of incoming data in recent years. In this paper, we present a recommendation system with the goal of helping knowledge workers in discovering useful new content. In particular, our system builds personalized user models based on file activities on enterprise network file servers. Our models use novel features that are derived from file metadata and user collaboration. Through extensive evaluation on real-world enterprise data, we demonstrate the effectiveness of our system with high precision and recall values. Unfortunately, our experiments reveal that per-user models are unable to handle heavy workloads. To address this limitation, we propose a novel optimization technique, active feature-based model selection, that predicts the user models that should be applied on each test file. Such a technique can reduce the classification time per file by as much as 23 times without sacrificing accuracy. We also show how this technique can be extended to improve the scalability exponentially at marginal cost of prediction accuracy, e.g., we can gain 169 times faster performance on an average across all shares by sacrificing 4% of F-score. |
| first_indexed | 2024-12-19T07:44:17Z |
| format | Article |
| id | doaj.art-f2f1deab002d4563b166be190d69e158 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-19T07:44:17Z |
| publishDate | 2016-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-f2f1deab002d4563b166be190d69e1582022-12-21T20:30:23ZengIEEEIEEE Access2169-35362016-01-01420421510.1109/ACCESS.2015.25130007368090Improving Scalability of Personalized Recommendation Systems for Enterprise Knowledge WorkersChetan Verma0Michael Hart1Sandeep Bhatkar2Aleatha Parker-Wood3Sujit Dey4Department of Electrical and Computer EngineeringMobile Systems Design Laboratory, University of California at San Diego, San Diego, CA, USA Symantec Research Labs, Mountain View, CA, USA Symantec Research Labs, Mountain View, CA, USA Symantec Research Labs, Mountain View, CA, USADepartment of Electrical and Computer EngineeringMobile Systems Design Laboratory, University of California at San Diego, San Diego, CA, USAEnterprise knowledge workers have been overwhelmed by the growing rate of incoming data in recent years. In this paper, we present a recommendation system with the goal of helping knowledge workers in discovering useful new content. In particular, our system builds personalized user models based on file activities on enterprise network file servers. Our models use novel features that are derived from file metadata and user collaboration. Through extensive evaluation on real-world enterprise data, we demonstrate the effectiveness of our system with high precision and recall values. Unfortunately, our experiments reveal that per-user models are unable to handle heavy workloads. To address this limitation, we propose a novel optimization technique, active feature-based model selection, that predicts the user models that should be applied on each test file. Such a technique can reduce the classification time per file by as much as 23 times without sacrificing accuracy. We also show how this technique can be extended to improve the scalability exponentially at marginal cost of prediction accuracy, e.g., we can gain 169 times faster performance on an average across all shares by sacrificing 4% of F-score.https://ieeexplore.ieee.org/document/7368090/Information RetrievalMachine LearningEnterpriseFile systems |
| spellingShingle | Chetan Verma Michael Hart Sandeep Bhatkar Aleatha Parker-Wood Sujit Dey Improving Scalability of Personalized Recommendation Systems for Enterprise Knowledge Workers IEEE Access Information Retrieval Machine Learning Enterprise File systems |
| title | Improving Scalability of Personalized Recommendation Systems for Enterprise Knowledge Workers |
| title_full | Improving Scalability of Personalized Recommendation Systems for Enterprise Knowledge Workers |
| title_fullStr | Improving Scalability of Personalized Recommendation Systems for Enterprise Knowledge Workers |
| title_full_unstemmed | Improving Scalability of Personalized Recommendation Systems for Enterprise Knowledge Workers |
| title_short | Improving Scalability of Personalized Recommendation Systems for Enterprise Knowledge Workers |
| title_sort | improving scalability of personalized recommendation systems for enterprise knowledge workers |
| topic | Information Retrieval Machine Learning Enterprise File systems |
| url | https://ieeexplore.ieee.org/document/7368090/ |
| work_keys_str_mv | AT chetanverma improvingscalabilityofpersonalizedrecommendationsystemsforenterpriseknowledgeworkers AT michaelhart improvingscalabilityofpersonalizedrecommendationsystemsforenterpriseknowledgeworkers AT sandeepbhatkar improvingscalabilityofpersonalizedrecommendationsystemsforenterpriseknowledgeworkers AT aleathaparkerwood improvingscalabilityofpersonalizedrecommendationsystemsforenterpriseknowledgeworkers AT sujitdey improvingscalabilityofpersonalizedrecommendationsystemsforenterpriseknowledgeworkers |