A holistic framework for user recommendation in social networks

Abstract

As social networks grow larger and larger, finding users of interest becomes an increasingly difficult task, making them great scenarios for the application of recommender systems. In this work, we perform an extensive evaluation of content-based, collaborative-based, and diffusion-based algorithms for user recommendation, and perform experiments on real datasets from Twitter. Previous research has shown the value in combining different recommendation algorithms, as each one has strengths and weaknesses. However, most have focused on specific classes of recommendation algorithms, or on naively combining them. In contrast, we present a holistic hybrid framework that simultaneously takes into account content-based, collaborative-based, diffusion-based, and user-based information. It learns how to combine different sources of evidence by using a Logistic Regression model. Our experiments show that our algorithm outperforms current state-of-the-art algorithms for user recommendation.