Analisis Sentimen Pengunjung Pasar Dayak Samarinda Menggunakan Algoritma Naive Bayes (Studi Kasus: Pasar Dayak Samarinda)

Buring, Julia Emakulata (2026) Analisis Sentimen Pengunjung Pasar Dayak Samarinda Menggunakan Algoritma Naive Bayes (Studi Kasus: Pasar Dayak Samarinda). S1 Teknik Informatika thesis, STMIK Widya Cipta Dharma.

	Text 2243098-S1-Jurnal.pdf Download (132kB)
	Text 2243098-S1-Teknik Informatika.pdf Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

Perkembangan teknologi informasi mendorong masyarakat untuk menyampaikan pendapat melalui platform digital, salah satunya Google Maps. Ulasan yang diberikan pengunjung dapat dimanfaatkan untuk mengetahui persepsi masyarakat terhadap suatu tempat, termasuk Pasar Dayak Samarinda. Namun, banyaknya jumlah ulasan menyebabkan proses analisis secara manual menjadi kurang efektif. Oleh karena itu, penelitian ini bertujuan untuk menganalisis sentimen pengunjung Pasar Dayak Samarinda berdasarkan ulasan Google Maps menggunakan algoritma Naive Bayes serta mengimplementasikan hasil analisis tersebut ke dalam website berbasis Flask. Penelitian ini menggunakan metode Cross Industry Standard Process for Data Mining (CRISP-DM) yang meliputi tahapan business understanding, data understanding, data preparation, modeling, evaluation, dan deployment. Data yang digunakan berasal dari ulasan pengunjung Pasar Dayak Samarinda di Google Maps. Tahap data preparation meliputi case folding, cleaning, repetition removal, normalisasi kata slang, tokenizing, stopword removal, dan stemming. Selanjutnya dilakukan ekstraksi fitur Menggunakan TF-IDF, penyeimbangan data dengan RandomOverSampler, serta klasifikasi menggunakan algoritma Multinomial Naive Bayes. Model dievaluasi menggunakan metrik Accuracy, Precision, Recall, dan F1-Score. Hasil penelitian menunjukkan bahwa model memperoleh nilai Accuracy sebesar 75,00%, Precision sebesar 71,27%, Recall sebesar 75,00%, dan F1-Score sebesar 70,41%. Hasil classification report menunjukkan bahwa model mampu mengklasifikasikan sentimen positif dengan sangat baik, sedangkan performa pada kelas netral dan negatif masih perlu ditingkatkan. Model yang telah dibangun kemudian diimplementasikan ke dalam website berbasis Flask sehingga pengguna dapat melakukan prediksi sentimen terhadap komentar baru serta melihat hasil analisis melalui antarmuka web. ============================================================ The development of information technology has encouraged people to express their opinions through digital platforms, one of which is Google Maps. Visitor reviews can be utilized to identify public perceptions of a particular place, including Pasar Dayak Samarinda. However, the large number of reviews makes manual analysis less effective. Therefore, this study aims to analyze the sentiments of Pasar Dayak Samarinda visitors based on Google Maps reviews using the Naive Bayes algorithm and to implement the analysis results in a Flask-based website. This study employed the Cross Industry Standard Process for Data Mining (CRISP-DM) methodology, which consists of business understanding, data understanding, data preparation, modeling, evaluation, and deployment. The data used in this study were obtained from visitor reviews of Pasar Dayak Samarinda on Google Maps. The data preparation stage included case folding, cleaning, repetition removal, slang word normalization, tokenization, stopword removal, and stemming. Feature extraction was then performed using TF-IDF, followed by data balancing using RandomOverSampler and classification using the Multinomial Naive Bayes algorithm. The model was evaluated using Accuracy, Precision, Recall, and F1-Score metrics. The results showed that the model achieved an Accuracy of 75.00%, Precision of 71.27%, Recall of 75.00%, and an F1-Score of 70.41%. The classification report indicated that the model performed very well in classifying positive sentiments, while its performance in identifying neutral and negative sentiments still needs improvement. The developed model was subsequently implemented in a Flask-based website, allowing users to predict the sentiment of new comments and view the analysis results through a web interface

Item Type:	Thesis (S1 Teknik Informatika)
Additional Information:	Pembimbing 1 : Dr. H. Nursobah, S.Kom., M.Kom., IPM., ASEAN Eng Pembimbing 2 : Rizky Zakariyya Rasyad, S.E., M.M
Uncontrolled Keywords:	Kata Kunci: Analisis Sentimen, Naive Bayes, TF-IDF, Flask, Google Maps, Pasar Dayak Samarinda
Subjects:	Q Science > QA Mathematics > QA76 Computer software
Divisions:	Teknik Informatika
Depositing User:	Ms Julia Emakulata Buring
Date Deposited:	06 Aug 2026 06:20
Last Modified:	06 Aug 2026 06:20
URI:	http://repository.wicida.ac.id/id/eprint/6385

Actions (login required)

View Item