Home // DATA ANALYTICS 2018, The Seventh International Conference on Data Analytics // View article

Stance Classification Using Political Parties in Tokyo Metropolitan Assembly Minutes

Authors:
Yasutomo Kimura
Minoru Sasaki

Keywords: Stance classification; Political party; Local assembly minutes.

Abstract:
Stance classification is an important component of argument mining. We focus on politicians' utterances in assembly minutes to classify political parties affiliation. This paper describes a novel stance classification task that classifies each politician's utterance for the politician's stance. Our task is to classify politicians into 20 political parties using their utterances in the Metropolitan Assembly minutes. Japanese assembly members are divided into many political parties in the local assembly. Our proposal is to apply several baseline methods to our novel dataset, which includes political parties in the Metropolitan Assembly minutes. In this paper, we define a political stance for a political party in Japan. We assess the difficulty of our dataset to evaluate several baseline methods, such as Support Vector machines (SVM), decision tree, random forest, and Naive Bayes.

Pages: 46 to 49

Copyright: Copyright (c) IARIA, 2018

Publication date: November 18, 2018

Published in: conference

ISSN: 2308-4464

ISBN: 978-1-61208-681-1

Location: Athens, Greece

Dates: from November 18, 2018 to November 22, 2018