Detection of Hoax Content on Social Media in Indonesia using a Levenshtein Distance Method

Frista Gifti Weddiningrum, Anang Kunaefi, Noor Wahyudi

2018

Abstract

Social media is a very supportive means to strengthen communication among fellow human beings. However, not all information disseminated through social media is a fact. There have been various cases of dissemination of news which are not facts or often called hoaxes. The development of anti-hoax technology has sprung up, but anti-hoax technology applied in the hoax detection system is still rarely found. In this study, Tf-Idf calculations were used to measure the weight of a word in a hoax document and the Levenshtein Distance (LD) method was used to measure the distance between words in a document. The application of the Levenshtein Distance Method in the Hoax Detection System has several steps, started with word pre-processing, followed by the Tf-Idf calculation phase, and then the calculation phase of the minimum inter-word distance using the Levenshtein Distance method. The results of the 0.0014 limit in the testing scenario have training data as many as 100 news indicated as hoaxes and 40 news as test data. The 0.0014 limit has a consistent value of Precision, Recall, and Accuracy

Paper Citation

in Harvard Style

Weddiningrum F., Kunaefi A. and Wahyudi N. (2018). Detection of Hoax Content on Social Media in Indonesia using a Levenshtein Distance Method. In Proceedings of the Built Environment, Science and Technology International Conference - Volume 1: BEST ICON, ISBN 978-989-758-414-5, pages 255-261. DOI: 10.5220/0008905600002481

in Bibtex Style

@conference{best icon18,
author={Frista Gifti Weddiningrum and Anang Kunaefi and Noor Wahyudi},
title={Detection of Hoax Content on Social Media in Indonesia using a Levenshtein Distance Method},
booktitle={Proceedings of the Built Environment, Science and Technology International Conference - Volume 1: BEST ICON,},
year={2018},
pages={255-261},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0008905600002481},
isbn={978-989-758-414-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the Built Environment, Science and Technology International Conference - Volume 1: BEST ICON,
TI - Detection of Hoax Content on Social Media in Indonesia using a Levenshtein Distance Method
SN - 978-989-758-414-5
AU - Weddiningrum F.
AU - Kunaefi A.
AU - Wahyudi N.
PY - 2018
SP - 255
EP - 261
DO - 10.5220/0008905600002481