Breaking the Communities: Characterizing community changing users using text mining and graph machine learning on Twitter

Abstract

The formation of majorities in public discussions often depends on individuals who shift their opinion over time. The detection and characterization of these type of individuals is therefore extremely important for political analysis of social networks. In this paper, we study changes in individual's affiliations on Twitter using natural language processing techniques and graph machine learning algorithms. In particular, we collected 9 million Twitter messages from 1.5 million users and constructed the retweet networks. We identified communities with explicit political orientation and topics of discussion associated to them which provide the topological representation of the political map on Twitter in the analyzed periods. With that data, we present a machine learning framework for social media users classification which efficiently detects "shifting users" (i.e. users that may change their affiliation over time). Moreover, this machine learning framework allows us to identify not only which topics are more persuasive (using low dimensional topic embedding), but also which individuals are more likely to change their affiliation given their topological properties in a Twitter graph.

Community pooling: LDA topic modeling in Twitter

“…A retweet graph was defined in terms of G = (N, E), where users were nodes N and retweets between them were edges E (Albanese et al, 2020). Since a user could retweet multiple times other user's tweets, the edges were weighted.…”

“…Considering that LDA allows multiple topics in one single document, having longer documents with denser co-occurence matrix has been shown to be beneficial (Alvarez-Melis & Saveski, 2016). Also, other works had shown that tweets belonging to users of one community in the retweet network were mostly of one or two topics (Albanese et al, 2020).…”

“…In this paper, we proposed a novel pooling techniques based on community detection on graphs. Previous works stated that LDA has problems with sparse word co-occurrence matrix (Ma et al, 2019) and showed that users in a community tweet mostly about one or two particular topics (Albanese et al, 2020). Based on these issues, we proposed a community pooling method which groups tweets whose authors belong to the same community on the retweet network, increasing the length of each document and reducing the total number of documents.…”