CEUR Workshop Proceedings. Proceedings of the International Workshop on Social Network Analysis using Formal Concept Analysis (SNAFCA 2015)
Social network analysis (SNA) is a multidisciplinary research area that has attracted many researchers from different disciplines such as Physics, Mathematics, Sociology, Biology and Computer Science, and has been studied according to different approaches and techniques. A social network is a dynamic structure (generally represented as a graph) of a set of entities/actors (nodes) together with links (edges) between them. The explosive growth of online social media has provided users with the opportunity to create and share digital content on a range hardly imaginable a few years ago. Indeed, massive participation has transformed online social networks into cores of social activity and a critical information vehicle. This is reflected by the number of news, opinions, and reviews that are constantly posted and discussed on these networks. The size and diversity of user generated content create an opportunity for identifying central and influential players, behavioral trends and user communities.
Nowadays social data analysts use a complicated mix of languages, methods and technologies for analyzing social networks services (SNS) data. In this article we describe approaches and technologies for extracting, analyzing and visualizing social data using Formal Concept Analysis Research Toolbox (FCART). Integrated process of analyzing SNS data with a set of research tools based on Formal Concept Analysis is considered with examples on datasets from Russian segment of LiveJournal.
Our research is focused on the study of social interactions of online community users, especially in business-oriented social network- ing services like LinkedIn or Habrahabr. The general aim of the work is to design methods for profiling of discussion participants within groups according to their interaction patterns. One of our goals is to make the approach independent from the language of communication, that is why we build our analysis on the comments graph and do not use information from the posts content. This paper suggest FCA based approach to pro-filing less active users for which not much data is available and statistical analysis is not applicable.