Section: New Results

Semantic and Temporal Analysis of Online Communities

Participant : Zide Meng.

The objective of the OCKTOPUS ANR project is to increase the potential social and economic benefit of the large and quickly growing amounts of user-generated content, by transforming it into useful knowledge. Since user communities are the basic of user-generated content sites, we start with community detection problem, which is a fundamental research point in social network analysis. Based on the preliminary experience from the previous year, we made several progress this year and published the results in international conferences, specifically:

• Topic based interested group detection:

  By analyzing a dataset extracted from the popular question answer site ”StackOverflow”, we proposed a heuristic method to enrich questions' tag. We also introduced a tag tree based model to extract topics from questions' tags, then we used the detected topics to label users in order to detect interest groups. We conducted experiments on the dataset and compared with related method. Results show that the proposed method is much simple and fast. This work has been published in [43] .

• Question Answer social media management

  We proposed a question answer social media system based on social network analysis and social media mining to manage the two main resources in question answer sites: users and contents. We also presented a vocabulary used to formalize both the level of interest and the expertise of users on topics. We tested QASM on a dataset extracted from the popular ”StackOverflow” site. We showed how the formalized knowledge is used to find relevant experts for a question. This work has been published in [95] .