(KLIK - S4) - Nursyamsyi, Hasan [2023-12-25]

Hasan, Firman Noor (2023) (KLIK - S4) - Nursyamsyi, Hasan [2023-12-25]. KLIK: Kajian Ilmiah Informatika dan Komputer, 4 (3). pp. 1788-1798. ISSN 2723-3898

Preview	Text 1. Cover.pdf Download (57kB) | Preview
Preview	Text 2. Editorial Team.pdf Download (498kB) | Preview
Preview	Text 3. Daftar Isi.pdf Download (13MB) | Preview
Preview	Text 4. Artikel.pdf Download (676kB) | Preview
Preview	Text 5. Bukti Hasil Turnitin.pdf Download (2MB) | Preview
Preview	Text 6. LOA - Nursyamsyi, Hasan.pdf Download (212kB) | Preview

Abstract

The Ministry of Home Affairs announced the implementation of an application to keep up with technological and information developments while utilizing digitalization in an effort to increase the efficiency of public services to the community in terms of population data under the name Digital Population Identity (IKD). The Digital Population Identity Application will represent identity data information in digital form. There have been 5 million users who have downloaded the application and around 33 thousand people have provided reviews regarding their satisfaction after using the application. However, implementing the Digital Population Identity application still has pros and cons. There are various user sentiments given based on reviews regarding their satisfaction after using the application. From this problem. The researcher tried to conduct sentiment classification research using the Naïve Bayes algorithm and Support Vector Machine using RapidMiner Studio to determine the public's response to their satisfaction with the Digital Population Identity application by pulling review data on the Digital Population Identity application. Sentiment in review data will be divided into positive sentiment and negative sentiment. The stages carried out in the research process are data collection, data labeling, data cleaning, word weighting with TF-IDF, SMOTE Upsampling, and Cross Validation to accommodate the two classification algorithms, apply model, and performance. As a result of the analysis process that has been carried out, the Support Vector Machine algorithm has quite good performance with an accuracy value of 80.46%, precision of 0.73, and recall of 0.96%. Meanwhile, Naïve Bayes has an accuracy value of 80.22%, precision of 0.73 and recall of 0.94. Both algorithms can carry out the classification process well in the analysis process in the Digital Population Identity application

Item Type:	Article
Subjects:	T Technology > T Technology (General)
Divisions:	Fakultas Teknik > Teknik Informatika
Depositing User:	Mr Firman Noor Hasan
Date Deposited:	09 Jan 2024 01:36
Last Modified:	09 Jan 2024 01:36
URI:	http://repository.uhamka.ac.id/id/eprint/30647

Actions (login required)

View Item