Analisis Sentimen Menggunakan Naive Bayes di PT Angkasa Pura I (Persero) Berdasarkan Ulasan pada Google Maps

Arofah, Muhimmatul (2025) Analisis Sentimen Menggunakan Naive Bayes di PT Angkasa Pura I (Persero) Berdasarkan Ulasan pada Google Maps. Project Report (Praktek Kerja Lapang). Universitas Pembangunan Nasional Veteran Jawa Timur.

	Text (Coover) 21083010055_Cover.pdf Download (2MB)
	Text (Bab 1) 21083010055_BAB 1.pdf Download (310kB)
	Text (Bab 2) 21083010055_BAB 2.pdf Restricted to Registered users only until June 2027. Download (398kB)
	Text (Bab 3) 21083010055_BAB 3.pdf Restricted to Registered users only until June 2027. Download (1MB)
	Text (Bab 4) 21083010055_BAB 4.pdf Download (186kB)
	Text (Daftar Pustaka) 21083010055_Daftar Pustaka.pdf Restricted to Registered users only until June 2027. Download (187kB)
	Text (Lampiran) 21083010055_Lampiran.pdf Download (520kB)

Abstract

User sentiment analysis is essential in evaluating public services, especially in the air transport sector, which plays a significant role in customer experience. PT Angkasa Pura I (Persero), as the manager of 15 airports in Indonesia, requires a deep understanding of user satisfaction through reviews available on digital platforms like Google Maps. This research aims to automatically analyze the sentiment of Google Maps reviews from 15 airports managed by PT Angkasa Pura I (Persero). The results of this analysis are expected to provide strategic insights for the company to enhance service quality according to user needs. The methodology involves data collection using web scraping techniques with Python libraries like BeautifulSoup and Selenium to extract reviews from Google Maps. The collected data is then processed through several preprocessing stages, including filtering, case folding, normalization, stopword removal, tokenization, stemming, and translation, to produce text ready for analysis. Data labeling is performed using the TextBlob library to categorize review sentiment as positive, neutral, or negative, with an initial result of 4,318 positive reviews, 443 neutral, and 652 negative. Subsequently, the Naive Bayes algorithm is applied to classify the sentiment with an accuracy reaching 83.54%. The classification results show 4,420 positive reviews, 523 neutral, and 443 negative, reflecting that most users are satisfied with the services provided. Further analysis of word frequency in reviews revealed commonly occurring words in both positive and negative reviews. This can be used to evaluate areas of service that need improvement and reinforcement. For example, positive reviews commonly include words such as “clean,” “good,” and “comfortable,” which reflect a positive experience, while negative reviews raise issues like “officer,” “terminal,” and “baggage,” which require the company’s attention. Based on these findings, PT Angkasa Pura I (Persero) can effectively set priorities for service improvements, enhance the user experience at the airport, and better manage public perception.

Item Type:

Monograph (Project Report (Praktek Kerja Lapang))

Contributors: