Virtual learning communities (VLCs) rethinking: influence on behavior modification—bullying detection through machine learning and natural language processing

Abstract

The current study attempts to investigate the influence of virtual learning communities (VLCs) on behavior modification, through the bullying paradigm, using natural language processing (NLP) techniques. The key question is whether individual learners that bully in their physical learning community (PLC) can be able to exhibit a behavior modification, if integrated in a VLC. Results indicate that the attempted "integration" could be a promising framework to behavior modification via a virtual community. Furthermore, machine learning is employed for the automatic detection of aggressive behavior that can facilitate the timely teacher's intervention, without him having to manually scan through the textual dataset. To the authors' knowledge, this is the first time such a linguistic and behavioral analysis for bullying detection is applied to VLCs. Another innovative challenge is the language targeted in the analysis, namely Greek.