POS-Tagging for informal language (study in Indonesian tweets)

This paper evaluates Part-of-Speech Tagging for the formal Indonesian language can be used for the tagging process of Indonesian tweets. In this study, we add five additional tags which reflect to social media attributes to the existing original tagset. Automatic POS tagging process is done by strat...

Full description

Saved in:
Bibliographic Details
Published in:Journal of physics. Conference series 2018-03, Vol.971 (1), p.12055
Main Authors: Suryawati, Endang, Munandar, Devi, Riswantini, Dianadewi, Abka, Achmad Fatchuttamam, Arisal, Andria
Format: Article
Language:English
Subjects:
Marking
Physics
Tags
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper evaluates Part-of-Speech Tagging for the formal Indonesian language can be used for the tagging process of Indonesian tweets. In this study, we add five additional tags which reflect to social media attributes to the existing original tagset. Automatic POS tagging process is done by stratified training process with 1000, 1600, and 1800 of annotated tweets. It shows that the process can achieve up to 66.36% accuracy. The experiment with original tagset gives slightly better accuracy (67.39%) than the experiment with five additional tags, but will lose important informations which given by the five additional tagset.POS-Tagging for Informal Language (Study in Indonesian Tweets).
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/971/1/012055