Please use this identifier to cite or link to this item: https://repositori.uma.ac.id/handle/123456789/27336
Title: Analisis Sentimen Komentar Pengunjung Terhadap Tempat Wisata Tjong A Fie Mansion Menggunakan Metode Naïve Bayes Classifier
Other Titles: Sentiment Analysis of Visitor Comments on Tjong A Fie Mansion Tourist Attraction Using the Naïve Bayes Classifier Method
Authors: Siregar, Erlina
metadata.dc.contributor.advisor: Lubis, Andre Hasudungan
Keywords: sentiment analysis;Naïve Bayes Classifier;tourist attraction;TF-IDF;text classification;klasifikasi teks;analisis sentimen
Issue Date: 2025
Publisher: Universitas Medan Area
Series/Report no.: NPM;188160066
Abstract: Penelitian ini bertujuan untuk menganalisis sentimen komentar pengunjung terhadap objek wisata Tjong A Fie Mansion di Kota Medan dengan menggunakan metode Naïve Bayes Classifier. Data diperoleh secara manual dari Google Maps sebanyak 100 komentar, yang kemudian melalui tahapan preprocessing meliputi case folding, tokenisasi, penghapusan stopword, dan stemming. Selanjutnya, dilakukan ekstraksi fitur menggunakan metode TF-IDF serta proses klasifikasi menggunakan algoritma Multinomial Naïve Bayes. Evaluasi kinerja model dilakukan dengan menggunakan confusion matrix. Hasil pengujian menunjukkan bahwa pembagian data pelatihan sebesar 80% dan data pengujian sebesar 20% menghasilkan akurasi tertinggi, yaitu sebesar 80% serta hasil sentimen 100% positif. Temuan ini menunjukkan bahwa metode Naïve Bayes mampu mengklasifikasikan komentar berbasis teks secara efektif dan efisien. Hasil analisis sentimen ini diharapkan dapat memberikan masukan bagi pengelola objek wisata dalam meningkatkan kualitas pelayanan, serta menjadi referensi dalam pengembangan sistem pendukung keputusan berbasis opini pengguna. This study aims to analyze the sentiment of visitor comments on the Tjong A Fie Mansion tourist attraction in Medan City using the Naïve Bayes Classifier method. A total of 100 comments were manually collected from Google Maps and underwent preprocessing stages, including case folding, tokenization, stopword removal, and stemming. Feature extraction was then performed using the TF-IDF method, followed by classification using the Multinomial Naïve Bayes algorithm. Model performance was evaluated using a confusion matrix. The test results showed that a data split of 80% for training and 20% for testing yielded the highest accuracy, reaching 80%, with a sentiment classification result of 100% positive. These findings indicate that the Naïve Bayes method can effectively and efficiently classify text-based comments. The sentiment analysis results are expected to provide input for tourism managers to improve service quality and serve as a reference for the development of user opinion-based decision support systems.
Description: 12 Halaman
URI: https://repositori.uma.ac.id/handle/123456789/27336
Appears in Collections:SP - Informatic Engineering

Files in This Item:
File Description SizeFormat 
188160066 - Erlina Siregar - Fulltext.pdfFulltext920.41 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.