Home // ICIMP 2020, The Fifteenth International Conference on Internet Monitoring and Protection // View article

A Lexicon Based Approach to Detect Extreme Sentiments

Authors:
Sebastião Pais
Irfan Tanoli
Miguel Albardeiro
João Cordeiro

Keywords: Natural Language Processing; Social Media; Extremism; Collective Radicalisation; Sentiment Analysis

Abstract:
Online social network platforms enable people freedom of expression to share their ideas, views, and emotions that could be negative or positive. Previous studies have investigated the user’s sentiments on such platforms to study the behaviour of people for different scenarios and purposes. The mechanism to collect information on public views attracted researchers by analyzing data from social networks and automatically classifying the polarity of public opinion(s) due to the use of concise language in posts or tweets. However, each cluster of tweet messages or posts focusing on a burst topic may constitute a potential threat to society and people. In this paper, we propose an unsupervised approach for automatic detection of people’s extreme sentiments on social networks. For this, our first task was automatically to build a standard lexicon consisting of extreme sentiments terms having high extreme positive and extreme negative polarity. With this new lexicon of extreme sentiments, our final task is to validate this lexicon, for which we developed an unsupervised approach for automatic detection of extreme sentiments, and we evaluated our performance on five different social networks and media datasets. This final task shows that, in these datasets, posts classified with negative sentiments, there are posts of extremely negative sentiments. On the other hand, in posts classified with positive sentiments, there are posts of extremely positive sentiments.

Pages: 1 to 6

Copyright: Copyright (c) IARIA, 2020

Publication date: September 27, 2020

Published in: conference

ISSN: 2308-3980

ISBN: 978-1-61208-804-4

Location: Lisbon, Portugal

Dates: from September 27, 2020 to October 1, 2020