Evaluating Public Anxiety for Topic-based Communities in Social Networks
Although individual anxiety evaluation has been well studied, few researches are developed to evaluate public anxiety of social network communities, which can benefit social network analysis tasks. However, we cannot simply average individual anxiety scores to evaluate a community's public anxiety, because following factors should be considered: (1) impacts from interpersonal relations on individuals anxiety levels (the Structural component); (2) topic-based discussions which reflect a community's anxiety status (the Topical component). In this paper, we initiate the study of evaluating public anxiety of Topic-based Social Network Communities (TSNC). We propose an evaluation framework to project the anxiety level of a TSNC into a score in [0, 1] range. We devise a cascading model to dynamically compute the individual anxiety scores using the Structural influence. We propose a probabilistic model to measure anxiety score of social network messages using a generalized user, and design a tree structure (MC-Tree) to effectively compute the anxiety score of a TSNC from the Topical component. For large communities, to avoid expensive real-time computing, we use a small sample to compute the public anxiety within given confidence interval. The effectiveness of our model are verified by precision and recall on Weibo and Twitter data sets.
Branch: CSE Domain: Data Mining
Developed In: Java