Statistical analysis of the social network and discussion threads in slashdot

Abstract

We analyze the social network emerging from the user comment activity on the website Slashdot. The network presents common features of traditional social networks such as a giant component, small average path length and high clustering, but differs from them showing moderate reciprocity and neutral assortativity by degree. Using Kolmogorov-Smirnov statistical tests, we show that the degree distributions are better explained by log-normal instead of power-law distributions. We also study the structure of discussion threads using an intuitive radial tree representation. Threads show strong heterogeneity and self-similarity throughout the different nesting levels of a conversation. We use these results to propose a simple measure to evaluate the degree of controversy provoked by a post.

Characterising Inter and Intra-Community Interactions in Link Streams Using Temporal Motifs

“…Name n m nodes edges Enron [9] 86978 1134990 employees e-mails Facebook [17] 45813 855542 users wall posts UC Irvine [13] 1899 59835 students messages Radoslaw [11] 167 82876 employees e-mails Debian [6] 34648 316569 users answers Digg [5] 30360 86203 users answers Linux Kernel Mailing List (LKML) 3 26885 1028233 users answers Slashdot [7] 51083 139789 users answers Table 1. Konect's networks Besides the Caen university dataset, we analysed a set of communication networks available on the Konect 4 website (see Table 1).…”

Predicting Instructor's Intervention in MOOC forums

“…The work presented in (Gómez et al, 2008;Liben-Nowell and Kleinberg, 2008;Kumar et al, 2010;Golub and Jackson, 2010;Wang et al, 2011;Aumayr et al, 2011) discuss characterizing threads using reply-graphs (often trees) and learning this structure. However, this representation is not natural for the MOOC domain where discussions are relatively more focused on the thread topic and are better organized using sections within the forums.…”

A Review of Existing Measures, Methods and Framework for Tracking Online Community in Social Network

“…It groups nodes into a cluster if the nodes are similar and then successively merges clusters until all nodes have been merged into a single remaining cluster. Techniques based on hierarchical clustering have been used to quantify the structure of community in documents [32], web pages, blogs, [33] and discussion groups [34]. Hierarchical clustering using such algorithms as in , results in a hierarchy (tree) being formed where the leaves of the tree are the nodes that are clustered.…”

“…In some cases human judgment, as reflected in the structuring of online communities and websites, can be inferred without additional data collection effort. For instance, researchers have used corroborating events, groups and categories inherent in the structure of online communities such as LiveJournal, DBLP and IMDB [34,46] to validate inferred community structure. User behaviour has also been studied in virtual communities .…”