Ali Mamdouh Elkahky
ABSTRACT
Recent online services rely heavily on automatic personalization to recommend relevant content to a large number of users. This requires systems to scale promptly to accommodate the stream of new users visiting the online services for the first time. In this work, we propose a content-based recommendation system to address both the recommendation quality and the system scalability. We propose to use a rich feature set to represent users, according to their web browsing history and search queries. We use a Deep Learning approach to map users and items to a latent space where the similarity between users and their preferred items is maximized. We extend the model to jointly learn from features of items from different domains and user features by introducing a multi-view Deep Learning model. We show how to make this rich-feature based user representation scalable by reducing the dimension of the inputs and the amount of training data. The rich user feature representation allows the model to learn relevant user behavior patterns and give useful recommendations for users who do not have any interaction with the service, given that they have adequate search and browsing history. The combination of different domains into a single model for learning helps improve the recommendation quality across all the domains, as well as having a more compact and a semantically richer user latent feature vector. We experiment with our approach on three real-world recommendation systems acquired from different sources of Microsoft products: Windows Apps recommendation, News recommendation, and Movie/TV recommendation. Results indicate that our approach is significantly better than the state-of-the-art algorithms (up to 49% enhancement on existing users and 115% enhancement on new users). In addition, experiments on a publicly open data set also indicate the superiority of our method in comparison with transitional generative topic models, for modeling cross-domain recommender systems. Scalability analysis show that our multi-view DNN model can easily scale to encompass millions of users and billions of item entries. Experimental results also confirm that combining features from all domains produces much better performance than building separate models for each domain.
- Fabian Abel, Qi Gao, Geert-Jan Houben, and Ke Tao. Twitter-based user modeling for news recommendations. In IJCAI'13, pages 2962--2966. Google Scholar
Digital Library
- Amr Ahmed, Abhimanyu Das, and Alexander J Smola. Scalable hierarchical multitask learning algorithms for conversion optimization in display advertising. In WSDM'14, pages 153--162. Google ScholarDigital Library
- Robert M Bell and Yehuda Koren. Improved neighborhood-based collaborative filtering. In KDD'13 CUP, 2007.Google Scholar
- Cheng Chu, Sang Kyun Kim, Yi-An Lin, YuanYuan Yu, Gary Bradski, Andrew Y Ng, and Kunle Olukotun. Map-reduce for machine learning on multicore. 2007.Google Scholar
- Adam Coates, Andrew Y Ng, and Honglak Lee. An analysis of single-layer networks in unsupervised feature learning. In AISTATS'11, pages 215--223.Google Scholar
- Abhinandan S Das, Mayur Datar, Ashutosh Garg, and Shyam Rajaram. Google news personalization: scalable online collaborative filtering. In WWW'07, pages 271--280. Google ScholarDigital Library
- Asela Gunawardana and Christopher Meek. Tied boltzmann machines for cold start recommendations. In RECSYS'08, pages 19--26. Google ScholarDigital Library
- John A Hartigan and Manchek A Wong. Algorithm as 136: A k-means clustering algorithm. Applied statistics, pages 100--108, 1979.Google Scholar
- Po-Sen Huang, Xiaodong He, Jianfeng Gao, Li Deng, Alex Acero, and Larry Heck. Learning deep structured semantic models for web search using clickthrough data. In CIKM'13, pages 2333--2338. Google ScholarDigital Library
- Piotr Indyk and Rajeev Motwani. Approximate nearest neighbors: towards removing the curse of dimensionality. In STOC'98, pages 604--613. Google ScholarDigital Library
- Joonseok Lee, Samy Bengio, Seungyeon Kim, Guy Lebanon, and Yoram Singer. Local collaborative ranking. In WWW'14, pages 85--96. Google ScholarDigital Library
- Bin Li, Qiang Yang, and Xiangyang Xue. Transfer learning for collaborative filtering via a rating-matrix generative model. In ICML'09, pages 617--624. Google ScholarDigital Library
- Ping Li, Trevor J Hastie, and Kenneth W Church. Very sparse random projections. In KDD'06, pages 287--296. Google ScholarDigital Library
- Greg Linden, Brent Smith, and Jeremy York. Amazon. com recommendations: Item-to-item collaborative filtering. Internet Computing, IEEE, 7(1):76--80, 2003. Google ScholarDigital Library
- Jiahui Liu, Peter Dolan, and Elin Rønby Pedersen. Personalized news recommendation based on click behavior. In IUI'10, pages 31--40. Google ScholarDigital Library
- Prem Melville, Raymond J Mooney, and Ramadass Nagarajan. Content-boosted collaborative filtering for improved recommendations. In AAAI'02. Google ScholarDigital Library
- Weike Pan, Evan Wei Xiang, Nathan Nan Liu, and Qiang Yang. Transfer learning in collaborative filtering for sparsity reduction. In AAAI'10, 2010.Google Scholar
- Al Mamunur Rashid, George Karypis, and John Riedl. Learning preferences of new users in recommender systems: an information theoretic approach. ACM SIGKDD Explorations Newsletter, 10(2):90--100, 2008. Google ScholarDigital Library
- Jasson DM Rennie and Nathan Srebro. Fast maximum margin matrix factorization for collaborative prediction. In ICML'05, pages 713--719. Google ScholarDigital Library
- Shaghayegh Sahebi and Peter Brusilovsky. Cross-domain collaborative recommendation in a cold-start context: The impact of user profile size on the quality of recommendation. In User Modeling, Adaptation, and Personalization, pages 289--295. Springer, 2013.Google Scholar
- Ruslan Salakhutdinov and Andriy Mnih. Bayesian probabilistic matrix factorization using markov chain monte carlo. In ICML'08, pages 880--887. Google ScholarDigital Library
- Ruslan Salakhutdinov, Andriy Mnih, and Geoffrey Hinton. Restricted boltzmann machines for collaborative filtering. In ICML'07, pages 791--798. Google ScholarDigital Library
- Badrul Sarwar, George Karypis, Joseph Konstan, and John Riedl. Item-based collaborative filtering recommendation algorithms. In WWW'01, pages 285--295. ACM, 2001. Google ScholarDigital Library
- Andrew I Schein, Alexandrin Popescul, Lyle H Ungar, and David M Pennock. Methods and metrics for cold-start recommendations. In SIGIR'02, pages 253--260. Google ScholarDigital Library
- Yang Song, Weiwei Cui, Shixia Liu, and Kuansan Wang. Online behavioral genome sequencing from usage logs: decoding the search behaviors. In WWW'14, pages 91--94. Google ScholarDigital Library
- Yang Song, Hongning Wang, and Xiaodong He. Adapting deep ranknet for personalized search. In WSDM'14, pages 83--92. Google ScholarDigital Library
- Shiliang Sun. A survey of multi-view machine learning. Neural Computing and Applications, 23(7--8):2031--2038, 2013.Google Scholar
- Jie Tang, Sen Wu, Jimeng Sun, and Hang Su. Cross-domain collaboration recommendation. In KDD'12, pages 1285--1293. Google ScholarDigital Library
- Bruce Thompson. Canonical correlation analysis. Encyclopedia of statistics in behavioral science, 2005.Google Scholar
- Aaron Van den Oord, Sander Dieleman, and Benjamin Schrauwen. Deep content-based music recommendation. In NIPS'13, pages 2643--2651.Google ScholarDigital Library
- Benjamin Van Durme and Ashwin Lall. Online generation of locality sensitive hash signatures. In ACL'10 Short Papers, pages 231--235. Google ScholarDigital Library
- Chong Wang and David M Blei. Collaborative topic modeling for recommending scientific articles. In KDD'11, pages 448--456. Google ScholarDigital Library
Index Terms
- A Multi-View Deep Learning Approach for Cross Domain User Modeling in Recommendation Systems
Recommendations
A Novel Cross-Domain Recommendation with Evolution Learning
In this “info-plosion” era, recommendation systems (or recommenders) play a significant role in finding interesting items in the surge of online digital activities and e-commerce. Several techniques have been widely applied for recommendation systems, but ...
Contrastive Multi-view Interest Learning for Cross-domain Sequential Recommendation
Cross-domain recommendation (CDR), which leverages information collected from other domains, has been empirically demonstrated to effectively alleviate data sparsity and cold-start problems encountered in traditional recommendation systems. However, ...
Cross-representation mediation of user models
Personalization is considered a powerful methodology for improving the effectiveness of information search and decision making. It has led to the dissemination of systems capable of suggesting relevant and personalized information (or items) to the users,...
Comments