Personalized recommendation on dynamic content using predictive bilinear models

By Wei Chu, Seung-Taek Park. 2009

280 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Information Systems, Management Science and Operations Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Yongfeng Zhang
2. Julian McAuley
3. Hongyuan Zha
4. Dacheng Tao
5. Luc Van Gool
6. Anton van den Hengel
7. Chunhua Shen
8. Jitendra Malik

In Information Systems

1. Yongfeng Zhang
2. Xu Chen
3. Lihong Li
4. Hongyuan Zha
5. J. Langford
6. Min Zhang
7. Julian McAuley
8. Yiqun Liu

In Management Science and Operations Research

1. Lihong Li
2. J. Langford
3. Nathan Kallus
4. Wei Chu
5. Thorsten Joachims
6. Robert E. Schapire
7. Chunqiu Zeng
8. Dietmar Jannach

Citations in the corpus

• Image-Based Recommendations on Styles and Substitutes (2015, background): “The system shares these characteristics with more common forms of recommender system, but does so on the basis of the appearance of the object, rather than metadata, reviews, or similar.”
• A contextual-bandit approach to personalized news article recommendation (2010, methodology): “We re-trained the bili near logistic regression model studied in [12] on Sept 2008 random t raffic data, using features zt,a constructed above.”
• Link Prediction via Matrix Factorization (2011, methodology): “In a collaborative filtering context, such bilinear models have been used in [8,37].”
• Collaborative competitive filtering: learning recommender using context of user choice (2011, methodology): “This model integrates both collaborative filtering [14] and content filtering [7].”
• CityTransfer (2018, methodology): “Based on all features in user and content profiles, Chu and Park [9] develop predictive bilinear regression models to provide accurate personalized recommendations of new items.”
• Daily-Aware Personalized Recommendation based on Feature-Level Time Series Analysis (2015, background): “The most early approaches for time-aware recommendation mainly focus on the dynamic modeling of user and item profiles [6, 2, 9].”
• Online Context-Aware Recommendation with Time Varying Multi-Armed Bandit (2016, background): “This problem makes recommender systems ineffective unless additional information about both items and users is collected [9][7].”
• News recommender systems - Survey and roads ahead (2018, methodology): “The two Front Page data sets were specifically constructed in a way so that an unbiased evaluation of recent explore-exploit based techniques (using contextual bandits) is possible [31, 106].”
• SCENE: a scalable two-stage personalized news recommendation system (2011, methodology): “For diversity evaluation, we choose [7] (Goo), [19] (ClickB), [5] (Bilinear) and [18] (Bandit) as the comparison baselines.”
• Personalizing LinkedIn Feed (2015, background): “This line of work is essentially close to the earlier personalized news article recommendation work [14], except that they used additional social network features like social voting, social tie strength etc. to model the activity relevance.”
• A case study of behavior-driven conjoint analysis on Yahoo!: front page today module (2009, methodology): “In our recent research [4], we observed that feature-based models yield higher CTR than conjoint models in offline analysis.”