On the quality of inferring interests from social neighbors

Abstract

This paper intends to provide some insights of a scientific problem: how likely one's interests can be inferred from his/her social connections -friends, friends' friends, 3-degree friends, etc? Is "Birds of a Feather Flocks Together" a norm? We do not consider the friending activity on online social networking sites. Instead, we conduct this study by implementing a privacy-preserving large distribute social sensor system in a large global IT company to capture the multifaceted activities of 30,000+ people, including communications (e.g., emails, instant messaging, etc) and Web 2.0 activities (e.g., social bookmarking, file sharing, blogging, etc). These activities occupy the majority of employees' time in work, and thus, provide a high quality approximation to the real social connections of employees in the workplace context. In addition to such "informal networks", we investigated the "formal networks", such as their hierarchical structure, as well as the demographic profile data such as geography, job role, self-specified interests, etc. Because user ID matching across multiple sources on the Internet is very difficult, and most user activity logs have to be anonymized before they are processed, no prior studies could collect comparable multifaceted activity data of individuals. That makes this study unique. In this paper, we present a technique to predict the inference quality by utilizing (1) network analysis and network autocorrelation modeling of informal and formal networks, and (2) regression models to predict user interest inference quality from network characteristics. We verify our findings with experiments on both implicit user interests indicated by the content of communications or Web 2.0 activities, and explicit user interests specified in user profiles. We demonstrate that the inference quality prediction increases the inference quality of implicit interests by 42.8%, and inference quality of explicit interests by up to 101%.

Learning relevance from heterogeneous social network and its application in online targeting

“…Examples include user interaction history during web search [3], user generated tags [21], and browsing behavior in quasi-social networks [19]. A recent work by Wen et al [24] studied users in social networks where there are multiple types of data. However, they assign each source an empirical weight when combining them and study the interest independently of the targeting task, while our model learns the weights from history data and can optimize the relevance measure towards different tasks.…”

“…Nevertheless their randomwalk-based approach suffer from the scalability issue in large social networks. Distinct with these works, we do not explicitly reply on the friendship link to propagate interests, as previous studies already suggest that explicit relationships are not good indicators of the nature of user interactions [4] and the quality of inferring user interests from their friends varies [24]. Instead, we use links to other entity nodes to propagate interests.…”

Large-scale behavioral targeting with a social twist

“…Engineers from Facebook [12] developed techniques to infer users' undeclared profiles (e.g., age, gender, profession) from their friends so that advertisers can precisely target more consumers. Scientists from academia also developed models to evaluate the quality of algorithms that derive one's interests from their social contacts [17,24]. However, most existing work is limited to the inference of "static" profiles such as age, gender, education.…”

Temporal Context-Aware Recommendation

“…Research efforts [12,15] have been undertaken to model users' interests to help them find interesting items by analyzing their historical behaviors. However, existing work [12,15,17] simply assumes that users prefer items based on their intrinsic interests, which may not be accurate in many social application scenarios. For example, when choosing a book to read or a movie to watch, the users are likely to prefer books/movies that interest them.…”