Article Per Year (5 Year)
p-Index From 2020 - 2025
0.408
P-Index
This Author published in this journals
All Journal IAES International Journal of Artificial Intelligence (IJ-AI)
Melani Puspasari, Hasna
Unknown Affiliation
Computer Science & IT Engineering

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Assessing public satisfaction of public service application using supervised machine learning Zharif Mustaqim, Ilham; Melani Puspasari, Hasna; Tri Utami, Avita; Syalevi, Rahmad; Ruldeviyani, Yova
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 13, No 2: June 2024
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v13.i2.pp1608-1618

Abstract

The COVID-19 pandemic has enormously affected the economic situation worldwide, including in Indonesia resulting in 30 million Indonesian tumbling into penury. The Ministry of Social Affairs initiated a program to distribute social assistance aimed at the poorest households. ‘Aplikasi Cek Bansos’ is a public service application that aims to validate their status towards the social assistance program. Understanding the public sentiment and factors affecting public satisfaction levels is crucial to be performed. The goal of this study is to perform a comparative study of supervised machine learning to learn the sentiment of the public and the dominant variable resulting in public satisfaction. Support vector machine, Naïve Bayes dan K-nearest neighbor (KNN) are performed to seek the highest accuracy. This experiment discovered that the KNN algorithm produced outstanding performance where the accuracy hit 99.21%. Sentiment prediction indicated negative perception as the majority covering 83.81%. Trigrams analysis is performed to learn themes affecting satisfaction levels toward the application. Negative themes are grouped into the following categories: App instability, hope for improvement, navigation issues, and low-quality content. Some recommendations are offered for the Ministry of Social Affairs and developers, to overcome negative feedback and enhance public satisfaction level towards the application.
Evaluation of Indonesia’s police public service platforms through sentiment and thematic analysis Melani Puspasari, Hasna; Zharif Mustaqim, Ilham; Tri Utami, Avita; Syalevi, Rahmad; Ruldeviyani, Yova
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 13, No 2: June 2024
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v13.i2.pp1596-1607

Abstract

The Indonesian national police (Polri) offer public services through mobile apps: Digital korlantas polri (DigiKorlantas) and samsat digital nasional (SIGNAL). Sentiment analysis gauges public perceptions, serving as a basis for e-government evaluation using user ratings and comments from app stores. Keyword relevance is assessed via feature extraction and Naïve Bayes classification. Thematic analysis is implemented using N-grams methods to identify the factors affecting the effectiveness based on user experiences. The accuracy of the model reaches 81.09% where it indicates a high performance. DigiKorlantas acquires slightly more negative reviews in comparation with positive reviews which are 51% and 49% respectively. In contrast, positive sentiment is dominant on SIGNAL which reach 58%, compared with negative sentiment that in 42%. N-grams reveal similar review patterns for both apps. Some of the solutions are Korlantas Polri should enhance the verification functionality with several techniques such as retinex algorithms or optical character recognition pipeline and increase the capacity of supporting server then releasing an updated version of application to address errors or bugs. This analysis can be alternative evaluation by the Polri to measure the success of the application and find out the continuous improvement of the process and the system.
Co-Authors Syalevi, Rahmad Tri Utami, Avita Yova Ruldeviyani Zharif Mustaqim, Ilham