Personalized Standard Deviations Improve the Baseline Estimation of Collaborative Filtering Recommendation
Baseline estimation is a critical component for latent factor-based collaborative filtering (CF) recommendations to obtain baseline predictions by evaluating global deviations for both users and items from personalized ratings. Classical baseline estimation presupposes that the user’s factual rating...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-07-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/10/14/4756 |
| _version_ | 1797562749525426176 |
|---|---|
| author | Zhenhua Tan Liangliang He Danke Wu Qiuyun Chang Bin Zhang |
| author_facet | Zhenhua Tan Liangliang He Danke Wu Qiuyun Chang Bin Zhang |
| author_sort | Zhenhua Tan |
| collection | DOAJ |
| description | Baseline estimation is a critical component for latent factor-based collaborative filtering (CF) recommendations to obtain baseline predictions by evaluating global deviations for both users and items from personalized ratings. Classical baseline estimation presupposes that the user’s factual rating range is the same as the system’s given rating range. However, from observations on real datasets of movie recommender systems, we found that different users have different actual rating ranges, and users can be classified into four kinds according to their personalized rating criterion, including normal, strict, lenient, and middle. We analyzed ratings’ distributions and found that the proportion of user ratings’ local standard deviation to the system’s global standard deviation is equal to that of the user’s actual rating range to the system’s rating range. We propose an improved and unified baseline estimation model based on the standard deviation’s proportion to alleviate the influence of classical baseline estimation’s limitation. We also apply the proposed baseline estimation model in existing latent factor-based CF recommendations and propose two instances. We performed experiments on full ratings of datasets by cross evaluations, including Flixster, Movielens (10 M), Movielens (latest small), FilmTrust, and MiniFilm. The results prove that the proposed baseline estimation model has better predictive accuracy than the classical model and is efficient in improving prediction performance for existing latent factor-based CF recommendations. |
| first_indexed | 2024-03-10T18:33:26Z |
| format | Article |
| id | doaj.art-574a750f67814a339562489c1d264341 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T18:33:26Z |
| publishDate | 2020-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-574a750f67814a339562489c1d2643412023-11-20T06:25:02ZengMDPI AGApplied Sciences2076-34172020-07-011014475610.3390/app10144756Personalized Standard Deviations Improve the Baseline Estimation of Collaborative Filtering RecommendationZhenhua Tan0Liangliang He1Danke Wu2Qiuyun Chang3Bin Zhang4Software College, Northeastern University, Shenyang 110819, ChinaCollege of Computer, National University of Defense Technology, Changsha 410073, ChinaSoftware College, Northeastern University, Shenyang 110819, ChinaSoftware College, Northeastern University, Shenyang 110819, ChinaSoftware College, Northeastern University, Shenyang 110819, ChinaBaseline estimation is a critical component for latent factor-based collaborative filtering (CF) recommendations to obtain baseline predictions by evaluating global deviations for both users and items from personalized ratings. Classical baseline estimation presupposes that the user’s factual rating range is the same as the system’s given rating range. However, from observations on real datasets of movie recommender systems, we found that different users have different actual rating ranges, and users can be classified into four kinds according to their personalized rating criterion, including normal, strict, lenient, and middle. We analyzed ratings’ distributions and found that the proportion of user ratings’ local standard deviation to the system’s global standard deviation is equal to that of the user’s actual rating range to the system’s rating range. We propose an improved and unified baseline estimation model based on the standard deviation’s proportion to alleviate the influence of classical baseline estimation’s limitation. We also apply the proposed baseline estimation model in existing latent factor-based CF recommendations and propose two instances. We performed experiments on full ratings of datasets by cross evaluations, including Flixster, Movielens (10 M), Movielens (latest small), FilmTrust, and MiniFilm. The results prove that the proposed baseline estimation model has better predictive accuracy than the classical model and is efficient in improving prediction performance for existing latent factor-based CF recommendations.https://www.mdpi.com/2076-3417/10/14/4756recommender systembaseline estimationcollaborative filteringstandard deviationpersonalization |
| spellingShingle | Zhenhua Tan Liangliang He Danke Wu Qiuyun Chang Bin Zhang Personalized Standard Deviations Improve the Baseline Estimation of Collaborative Filtering Recommendation Applied Sciences recommender system baseline estimation collaborative filtering standard deviation personalization |
| title | Personalized Standard Deviations Improve the Baseline Estimation of Collaborative Filtering Recommendation |
| title_full | Personalized Standard Deviations Improve the Baseline Estimation of Collaborative Filtering Recommendation |
| title_fullStr | Personalized Standard Deviations Improve the Baseline Estimation of Collaborative Filtering Recommendation |
| title_full_unstemmed | Personalized Standard Deviations Improve the Baseline Estimation of Collaborative Filtering Recommendation |
| title_short | Personalized Standard Deviations Improve the Baseline Estimation of Collaborative Filtering Recommendation |
| title_sort | personalized standard deviations improve the baseline estimation of collaborative filtering recommendation |
| topic | recommender system baseline estimation collaborative filtering standard deviation personalization |
| url | https://www.mdpi.com/2076-3417/10/14/4756 |
| work_keys_str_mv | AT zhenhuatan personalizedstandarddeviationsimprovethebaselineestimationofcollaborativefilteringrecommendation AT lianglianghe personalizedstandarddeviationsimprovethebaselineestimationofcollaborativefilteringrecommendation AT dankewu personalizedstandarddeviationsimprovethebaselineestimationofcollaborativefilteringrecommendation AT qiuyunchang personalizedstandarddeviationsimprovethebaselineestimationofcollaborativefilteringrecommendation AT binzhang personalizedstandarddeviationsimprovethebaselineestimationofcollaborativefilteringrecommendation |