Title: A novel approach for visual data mining in Twitter micro-blogging platform

Authors: Manolis Maragoudakis; Ioannis Markou

Addresses: Department of Information and Communication Systems Engineering, University of the Aegean, Karlovassi, Samos 83200, Greece ' Department of Information and Communication Systems Engineering, University of the Aegean, Karlovassi, Samos 83200, Greece

Abstract: In today's world of data dominance, social networking websites and especially micro-blogging platforms, form the largest share in current unstructured textual data. If the proper tools, such as sentiment analysis and topic modelling are applied to that data, valuable information would be produced. That information in turn could offer insights from understanding market trends to interpreting social phenomena. A question that comes to mind is: How can we use that information to enhance knowledge discovery and results readability? The answer is data visualisation. The purpose of this paper is to present a tool that was designed and developed to make use of network analysis algorithms in order to visualise social networking data. The goal of implementing this system is to create visual patterns from social media that leads to knowledge discovery and readable analysis results.

Keywords: data mining; data visualisation; textual data analysis; sentiment analysis; topic modelling; social media; Twitter; data cleaning; data reprocessing; microblogging; social networking; social network analysis; SNA; knowledge discovery; results readability.

DOI: 10.1504/IJSNM.2016.082647

International Journal of Social Network Mining, 2016 Vol.2 No.3, pp.267 - 292

Accepted: 02 Jul 2016
Published online: 04 Mar 2017 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article