IPTEK Journal of Science
Vol 2, No 1 (2017)

Tax Complaints Classification on Twitter Using Text Mining

Prita Dellia (Institut Teknologi Sepuluh Nopember, Surabaya, 60111, Indonesia)
Aris Tjahyanto (Institut Teknologi Sepuluh Nopember, Surabaya, 60111, Indonesia)



Article Info

Publish Date
01 May 2017

Abstract

Twitter growth and utilization encourage the emergence of limitless textual information so that people can express their complaints easily This leads the Directorate General of Taxation uses twitter to deal with tax complaints faced by the community. However, the messages on twitter can contain any information, either the tax complaint or not. This will cause difficulties in handling complaints process. It is important to automatically identify so tax complaint handling can be done effectively and efficiently. Given these problems, it is necessary to do the twitter tax complaint classification with the support of text mining. There are several methods of classification such as Naïve Bayes classifiers, Support Vector Machine (SVM) and Decision Tree. This research aims to classify the tax complaint on twitter automatically by using text mining. The experimental results show the value of f-measure of SVM, Naïve Bayes and Decision Tree, respectively, are 89.3%, 85.6% and 76.9%

Copyrights © 2017






Journal Info

Abbrev

jos

Publisher

Subject

Description

IPTEK Jounal of Science (E-ISSN: 2337-8530) is an academic journal on the issued related to natural science, mathematics, and statistics. Published actually in March, June, September, and December. It is open to all scientist, researchers, education practitioners, and other scholar. Therefore this ...