Your Tweets Reveal What You Like: Introducing Cross-media Content Information into Multi-domain Recommendation

Cold start is a challenging problem in recommender systems. Many previous studies attempt to utilize extra information from other platforms to alleviate the problem. Most of the leveraged information is on-topic, directly related to users' preferences in the target domain. Thought to be unrelated, users' off-topic content information (such as user tweets) is usually omitted. However, the off-topic content information also helps to indicate the similarity of users on their tastes, interests, and opinions, which matches the underlying assumption of Collaborative Filtering (CF) algorithms. In this paper, we propose a framework to capture the features from user's off-topic content information in social media and introduce them into Matrix Factorization (MF) based algorithms. The framework is easy to understand and flexible in different embedding approaches and MF based algorithms. To the best of our knowledge, there is no previous study in which user's off-topic content in other platforms is taken into consideration. By capturing the cross-platform content including both on-topic and off-topic information, multiple algorithms with several embedding learning approaches have achieved significant improvements in rating prediction on three datasets. Especially in cold start scenarios, we observe greater enhancement. The results confirm our suggestion that off-topic cross-media information also contributes to the recommendation.

Request-Driven Cross-Media Content Adaptation Technique

Devices, standards and software develop rapidly, but still often independently of each other. This creates problems in terms of content suitability on various devices. Also, in mobile environment, user and system-level applications must execute subject to a variety of resource constraints. In order to deal with these constraints, content adaptation is required. In this chapter, we justify the need of distributed cross media content adaptation and the potential of utilizing Web Services as the adaptation providers. We introduce request-driven context to complement constraint-driven and utility-driven approaches. We describe the request context mapping and propose a novel path’s determination scheme for determining the optimal service proxies to facilitate the adaptation tasks. To better illustrate the disjoint portions in content passing between service proxies, two communication models were associated. Then, within Web Services, we explain the related protocols and socket connection between adaptation’s services. We conclude with discussion regarding the strengths of the proposed architecture.