HATE SPEECH PREDICTION USING K-MEANS ALGORITHM

PRATAMA, LIM, ALEXANDRE NOVANDRYAN (2021) HATE SPEECH PREDICTION USING K-MEANS ALGORITHM. Other thesis, Universitas Katholik Soegijapranata Semarang.

[img]
Preview
Text
17.K1.0010-LIM, ALEXANDRE NOVANDRYAN PRATAMA-COVER_a.pdf

Download (1MB) | Preview
[img]
Preview
Text
17.K1.0010-LIM, ALEXANDRE NOVANDRYAN PRATAMA-BAB I_a.pdf

Download (244kB) | Preview
[img] Text
17.K1.0010-LIM, ALEXANDRE NOVANDRYAN PRATAMA-BAB II_a.pdf
Restricted to Registered users only

Download (222kB)
[img]
Preview
Text
17.K1.0010-LIM, ALEXANDRE NOVANDRYAN PRATAMA-BAB III_a.pdf

Download (251kB) | Preview
[img]
Preview
Text
17.K1.0010-LIM, ALEXANDRE NOVANDRYAN PRATAMA-BAB IV_a.pdf

Download (1MB) | Preview
[img]
Preview
Text
17.K1.0010-LIM, ALEXANDRE NOVANDRYAN PRATAMA-BAB V_a.pdf

Download (1MB) | Preview
[img]
Preview
Text
17.K1.0010-LIM, ALEXANDRE NOVANDRYAN PRATAMA-BAB VI_a.pdf

Download (205kB) | Preview
[img]
Preview
Text
17.K1.0010-LIM, ALEXANDRE NOVANDRYAN PRATAMA-DAPUS_a.pdf

Download (226kB) | Preview
[img]
Preview
Text
17.K1.0010-LIM, ALEXANDRE NOVANDRYAN PRATAMA-LAMP_a.pdf

Download (862kB) | Preview

Abstract

In the social media nowadays, there are lots of posts that shares a story about the user on something. Either it’s to share a moment, or an opinion. So does freedom speech goes, not a few who abuse the freedom speech to take down others. Mostly those who abuse the freedom speech use hate speech to make the interlocutor feels uncomfortable. This research is discusses about the usage of data mining algorithm and twitter data to predict hate speech in a post or a tweet. The K-Means that being used in this research is to define the Hate Speech in the dataset. The k is set to 2 to differentiate the first cluster is Hate Speech, and the second is Not Hate Speech. The final results offered is in the form of a percentage of accuracy and comparison of the amount of data. From various comparison of data, the highest accuracy that being achieved is 80% followed by 66,7%, etc. However, by the results shown that different methods may varies different results in accuracy. But in overall the most stable results is by using N-Gram Tri-gram.

Item Type: Thesis (Other)
Subjects: 000 Computer Science, Information and General Works > 004 Data processing & computer science
Divisions: Faculty of Computer Science > Department of Informatics Engineering
Depositing User: mr AM. Pudja Adjie Sudoso
Date Deposited: 15 Oct 2021 02:14
Last Modified: 15 Oct 2021 02:14
URI: http://repository.unika.ac.id/id/eprint/27125

Actions (login required)

View Item View Item