Studying Positive Speech on Twitter

TL;DR

This paper empirically investigates positive speech on Twitter, revealing it constitutes less than 1% of data and evaluating automated methods for its detection in a conflict context.

Contribution

It introduces automated approaches for detecting positive speech on Twitter and compares their effectiveness in a conflict-prone environment.

Findings

Positive speech is less than 1% of Twitter data.

Unsupervised and supervised methods have distinct benefits and challenges.

Empirical evidence supports the feasibility of automated detection.

Abstract

We present results of empirical studies on positive speech on Twitter. By positive speech we understand speech that works for the betterment of a given situation, in this case relations between different communities in a conflict-prone country. We worked with four Twitter data sets. Through semi-manual opinion mining, we found that positive speech accounted for < 1% of the data . In fully automated studies, we tested two approaches: unsupervised statistical analysis, and supervised text classification based on distributed word representation. We discuss benefits and challenges of those approaches and report empirical evidence obtained in the study.