Improving Users' Acceptance in Recommender System

Abstract

Personalized recommender systems aim to push only the relevant items and information directly to the users without requiring them to browse through millions of web resources. The challenge of these systems is to achieve a high user acceptance rate on their recommendations. Collaborative filtering is a method of increasing user? acceptance towards recommendation (filtering) about the interests of a user by collecting preferences or taste information from many users (collaborating). In this thesis, we focus on improving user?s acceptance by collaborative filtering of fusing different popular user-generated data types data to improve users' acceptance like social tagging and rating data, cross domain data and social trust data.

First, we study the problem of increasing the user?s acceptance using social tagging and rating data. We show that ternary relationships such as users-items-ratings, or users-items-tags, are insufficient to increase user? acceptance towards recommendations. Instead, we model the quaternary relationship among users, items, tags and ratings as a 4-order tensor and cast the recommendation problem as a multi-way latent semantic analysis problem. A unified framework for user recommendation, item recommendation, tag recommendation and item rating prediction is proposed. Next, we study the problem of increasing the user?s acceptance using cross domain data, which enables more accurate recommendation by utilizing the knowledge in the other domain. Finally, we study the problem of increasing the user?s acceptance using social trust data. We show that the complex interaction between user interests and the social relationship over time is important to increase the user?s acceptance toward recommendation, which is ignored by existing recommender systems model.