Collective Personal Profile Summarization with Social Networks

Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, pages 715–725, Seattle, Washington, USA, 18-21 October 2013. c⃝2013 Association for Computational Linguistics Collective Personal Profile Summarization with Social Networks Zhongqing Wang, Shoushan Li*, Kong Fang, and Guodong Zhou Natural Language Processing Lab, School of Computer Science and Technology Soochow University, Suzhou, 215006, China {wangzq.antony, shoushan.li}@gmail.com, {kongfang, gdzhou}@suda.edu.cn Abstract Personal profile information on social media like LinkedIn.com and Facebook.com is at the core of many interesting applications, such as talent recommendation and contextual advertis- ing. However, personal profiles usually lack or- ganization confronted with the large amount of available information. Therefore, it is always a challenge for people to find desired information from them. In this paper, we address the task of personal profile summarization by leveraging both personal profile textual information and so- cial networks. Here, using social networks is motivated by the intuition that, people with similar academic, business or social connections (e.g. co-major, co-university, and co- corporation) tend to have similar experience and summaries. To achieve the learning process, we propose a collective factor graph (CoFG) model to incorporate all these resources of knowledge to summarize personal profiles with local textual attribute functions and social connection factors. Extensive evaluation on a large-scale dataset from LinkedIn.com demonstrates the effective- ness of the proposed approach.* 1 Introduction Web 2.0 has empowered people to actively interact with each other, forming social networks around mutually interesting information and publishing a large amount of useful user-generated content (UGC) online (Lappas et al., 2011; Tan et al., 2011). One popular and important type of UGC is the personal profile, where people post detailed * Corresponding author information on online portals about their education, experiences and other personal information. Social websites like Facebook.com and LinkedIn.com have created a viable business as profile portals, with the popularity and success partially attributed to their comprehensive personal profiles. Generally, online personal profiles provide val- uable resources for businesses, especially for hu- man resource managers to find talents, and help people connect with others of similar backgrounds (Yang et al., 2011a; Guy et al., 2010). However, as there is always large-scale information of experi- ence and education fields, it is hardly for us to find useful information from the profile. Therefore, it is always a challenge for people to find desired in- formation from them. For this regard, it is highly desirable to develop reliable methods to generate a summary of a person through his profile automati- cally. To the best of our knowledge, this is the first re- search that explores automatic summarization of personal profiles in social media. A straightfor- ward approach is to consider personal profile summarization as a traditional document summari- zation problem, which treating each personal pro- file independently and generate a summary for each personal profile individually. For example, the well-known extraction and ranking approaches (e.g. PageRank, HITS) extract a certain amount of important sentences from a document according to some ranking measurements to form a summary (Wan and Yang, …

Content truncated to 500 words