Article Per Year (5 Year)
p-Index From 2021 - 2026
0.61
P-Index
This Author published in this journals
All Journal (JELIKU) Jurnal Elektronik Ilmu Komputer Udayana JUITA : Jurnal Informatika Journal of Language Education Studies and Applied Linguistics (JoLESAL)
Mahendra, Ida Bagus
Unknown Affiliation
Humanities Computer Science & IT Education Languange, Linguistic, Communication & Media

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : JUITA : Jurnal Informatika

 1 
Comparison of Multinomial, Bernoulli, and Gaussian Naïve Bayes for Complaint Classification in Pro Denpasar Application Mahendra, Ida Bagus; Sunarya, I Made Gede; Wirawan, I Made Agus
JUITA: Jurnal Informatika JUITA Vol. 13 Issue 1, March 2025
Publisher : Department of Informatics Engineering, Universitas Muhammadiyah Purwokerto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30595/juita.v13i1.24828

Abstract

Pelayanan Rakyat Online Denpasaror PRO Denpasar is a Denpasar City Government application intended as a public service mall to support Denpasar to become a smart city. This application was built since 2014 and is actively used in channeling public complaints so optimization is continuously needed to increase the efficiency of application use. Application optimization is carried out by developing decision support tools to determine complaint categories that are still done manually. The application of the properartificial intelligence method can be used as a solution in classifying complaint categories to become a decision support tool for operators. This study compares three classification methods including multinomial naïve bayes, bernoulli naïve bayes and gaussian naïve bayes by applying TF-IDF feature extraction to determine the best complaint category classification method. Based on eight comparison scenario results by applying a comparison of 25%, 50%, 75% and 100% of complaint descriptions with 5-fold cross validation and 10-fold cross validation, it was found that the multinomial naïve Bayes method provided the best result in seven combined comparisons involving the test parameters accuracy, precision, recall, f1-score and processing time.
Co-Authors Agus Muliantara Hermawan, Gede Satya I Made Gede Sunarya M.Cs S.Kom I Made Agus Wirawan . Patmi, Ni Putu Dita Candra Pramana Pradnyana, Ida Bagus Made