Article Per Year (5 Year)
p-Index From 2021 - 2026
0
P-Index
This Author published in this journals
All Journal Jurnal Mantik
Satrio Yudho Pangestu
Universitas AMIKOM Yogyakarta
Computer Science & IT Economics, Econometrics & Finance Languange, Linguistic, Communication & Media

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Algoritma Support Vector Machine Untuk Klasifikasi Sikap Politik Terhadap Partai Politik Indonesia: Algoritma Support Vector Machine Untuk Klasifikasi Sikap Politik Terhadap Partai Politik Indonesia Satrio Yudho Pangestu; Yuli Astuti; Lilis Dwi Farida
Jurnal Mantik Vol. 3 No. 1 (2019): May: Manajemen dan Informatika
Publisher : Institute of Computer Science (IOCS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (542.801 KB)

Abstract

The use of social media that is increasingly easy and affordable becomes a new forum for Indonesian people to express their thoughts freely. Included in the preparation period for a democratic party held every five years. The public can freely believe through the social media they have, especially through twitter. People who come from different backgrounds often provide opinions that can lead to pros and cons. It can be used as feedback on political parties that carry presidential, vice-presidential, and successful team candidates so that they will be useful in potential assessments and can be used for better purposes. Sentiment analysis is done by sorting data from Twitter which is an opinion on political parties and the executive candidates they carry. The data is divided into 2 categories, positive and negative categories. The methods that will be used for sentiment analysis include preprocessing, word staining with TF-IDF, and making a classification model with the Support Vector Machine and K-Fold Cross Validation approach to test the accuracy of the model. The result of making a classification model is Support Vector Machine with training data of 900 to get 86% accuracy and testing using 10-Fold Cross Validation get an average accuracy rate of 71% with an error rate of 29%
Co-Authors LILIS DWI FARIDA Yuli Astuti