Stance detection in online discussions

TL;DR

This paper presents a stance detection system that classifies online comments as in favor or against a target, adapting a maximum entropy classifier from English tweets to Czech news comments.

Contribution

It introduces an adaptation of a stance detection system from English to Czech, utilizing surface, sentiment, and domain-specific features with a maximum entropy classifier.

Findings

Effective adaptation to Czech news comments

Utilizes diverse feature types for stance detection

Demonstrates cross-lingual applicability of the approach

Abstract

This paper describes our system created to detect stance in online discussions. The goal is to identify whether the author of a comment is in favor of the given target or against. Our approach is based on a maximum entropy classifier, which uses surface-level, sentiment and domain-specific features. The system was originally developed to detect stance in English tweets. We adapted it to process Czech news commentaries.