Farikh Fadhil
Universitas Muhammadiyah Sukabumi

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

Found 1 Documents
Search

Implementation Of the Naive Bayes Algorithm for Analysis Sentiment of Alun-Alun as A Public Open Space in Sukabumi Regency Farikh Fadhil; Asriyanik Asriyanik; Winda Apriandari
Bit (Fakultas Teknologi Informasi Universitas Budi Luhur) Vol 20, No 2 (2023): SEPTEMBER 2023
Publisher : Universitas Budi Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36080/bit.v20i2.2434

Abstract

Sukabumi Regency is one of the regencies in West Java that has various facilities and infrastructure, including public open spaces, which serve as beneficial places for the daily activities and needs of the general public. One of these spaces is the town squares, namely Alun-Alun Cisaat, Alun-Alun Palabuhanratu, Alun-Alun Jampang Kulon, Alun-Alun Purabaya, and Alun-Alun Cicurug. However, there are still town squares in Sukabumi Regency that are poorly maintained, resulting in negative articles or opinions from visitors regarding these town squares. Therefore, it is necessary for the local government to undertake development or evaluation to further improve the public open spaces in Sukabumi Regency. Before carrying out the development and evaluation, it is essential to gather information, data, or conduct analysis related to public opinions on these public open spaces. One method is by performing sentiment analysis. This research conducts a public sentiment analysis on the town squares as public open spaces in Sukabumi Regency, utilizing the Naïve Bayes algorithm for sentiment analysis. The data used for analysis were obtained from visitor reviews of these places from the period of 2019 to 2023 on the Google Maps website. A total of 2698 sentiment data points were collected, consisting of 2254 positive sentiment data and 444 negative sentiment data. The algorithm used in the created model achieved an accuracy rate of 92%, precision value of 90%, recall value of 53%, and an F1-score of 67%. The frequency of words obtained from the sentiment analysis revealed the top 5 most frequently mentioned words based on their sentiment class. The top 5 positive sentiment words are excellent, good, clean, hang out, and pleasant. On the other hand, the top 5 negative sentiment words are garbage, dirty, slum, traffic jam, and disorganized. The results of this research, specifically related to sentiment analysis, are expected to assist the local government, especially Regional Office of Land and Spatial Planning in Sukabumi Regency, by providing valuable information as a reference or recommendation for the future development and evaluation of public open spaces in Sukabumi Regency.