Stance Classification Towards Political Figures on Blog Writing

Rini Jannati, Rahmad Mahendra, Cakra Wishnu Wardhana, Mirna Adriani

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

9 Citations (Scopus)

Abstract

In this paper, we present new application of stance detection task on politic domain. Our goal is to determine whether the writer of the blog article is on the position supporting a political figure to compete and win in a general election event, for example a candidate of President in the Presidential election. We performed the experiment using five different case studies. We examined three baseline machine learning models using combination of n-gram, sentiment lexicon, orthography, and word embedding features. The highest macro-average F1 score was achieved by model trained on Support Vector Machine classifier using a combination of word2vec and unigram features, which is 63,54%.

Original languageEnglish
Title of host publicationProceedings of the 2018 International Conference on Asian Language Processing, IALP 2018
EditorsMinghui Dong, Moch. Bijaksana, Herry Sujaini, Arif Bijaksana Putra Negara, Ade Romadhony, Fariska Z. Ruskanda, Elvira Nurfadhilah, Lyla Ruslana Aini
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages96-101
Number of pages6
ISBN (Electronic)9781728111766
DOIs
Publication statusPublished - 2 Jul 2018
Event22nd International Conference on Asian Language Processing, IALP 2018 - Bandung, Indonesia
Duration: 15 Nov 201817 Nov 2018

Publication series

NameProceedings of the 2018 International Conference on Asian Language Processing, IALP 2018

Conference

Conference22nd International Conference on Asian Language Processing, IALP 2018
Country/TerritoryIndonesia
CityBandung
Period15/11/1817/11/18

Keywords

  • Blog
  • Political Text Analytics
  • Stance Detection
  • Text Classification

Fingerprint

Dive into the research topics of 'Stance Classification Towards Political Figures on Blog Writing'. Together they form a unique fingerprint.

Cite this