Analisis Ulasan Pengunjung Akomodasi Wisata di Bali Menggunakan Latent Dirichlet Allocation dengan Pendekatan Topic-Based Sentiment Analysis

Salsabila, Fadhila (2024) Analisis Ulasan Pengunjung Akomodasi Wisata di Bali Menggunakan Latent Dirichlet Allocation dengan Pendekatan Topic-Based Sentiment Analysis. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003201131-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (4MB) | Request a copy

Abstract

Provinsi Bali menyumbang 50% dari total pendapatan devisa Indonesia di sektor pariwisata. Kesuksesan pariwisata sangat bergantung pada kualitas dan ketersediaan akomodasi wisata. Akomodasi wisata tidak hanya sebagai tempat menginap, tetapi juga memainkan peran penting dalam menciptakan kesan dan harapan bagi para wisatawan. Namun, pandemi Covid-19 menyebabkan jumlah kunjungan pada akomodasi wisata menurun. Oleh karena itu, pihak akomodasi wisata dapat memperhatikan feedback yang diberikan pengunjung, khususnya pada platform digital, TripAdvisor. Analisis pada penelitian ini diawali oleh esktraksi topik pada periode pre-Covid 19 dan during-Covid 19 menggunakan Latent Dirichlet Allocation (LDA). Pada kedua periode diperoleh 3 topik utama mengenai suasana dan pengalaman liburan, akesibilitas dan value, serta fasilitas dan layanan dengan coherence score pada masing-masing periode sebesar 0,5491 dan 0,4982. Analisis sentimen dilakukan pada setiap topik (topic-based sentiment analysis) menggunakan metode Naïve Bayes Classifier (NBC), baik menggunakan oversampling dengan SMOTE maupun tanpa SMOTE. Perbandingan metode tersebut menggunakan Stratified 10-Fold Cross Validation dengan kriteria kebaikan klasifikasi Area Under Curve (AUC) menunjukkan bahwa hasil performa NBC dengan SMOTE lebih baik ditunjukkan dengan nilai AUC diatas 80% untuk data testing di setiap topik pada masing-masing periode. Penelitian diakhiri dengan visualisasi menggunakan Lexical Salience-Valence Analysis (LSVA) Quadrant yang didasarkan pada perhitungan word salience dan word valence untuk melihat tingkat positivitas atau negatifitas serta mengamati perubahan dan pergeseran preferensi pengalaman pengunjung dengan adanya pandemi.
===================================================================================================
Bali Province accounts for 50% of Indonesia's total foreign exchange earnings in the tourism sector. The success of tourism depends heavily on the quality and availability of tourist accommodation. Tourist accommodation is not only a place to stay, but it also plays an important role in creating impressions and expectations for tourists. However, the Covid-19 pandemic has caused the number of visits to tourist accommodation to decrease. Therefore, tourist accommodations can pay attention to the feedback given by visitors, especially on the digital platform, TripAdvisor. The analysis in this study began with topic extraction in the pre-Covid 19 and during-Covid 19 periods using Latent Dirichlet Allocation (LDA). In both periods, 3 main topics were obtained regarding the atmosphere and experience of the holiday, accessibility and value, as well as facilities and services with a coherence score of 0.5491 and 0.4982 in each period. Sentiment analysis was carried out on each topic (topic-based sentiment analysis) using the Naïve Bayes Classifier (NBC) method, both using oversampling with SMOTE and without SMOTE. The comparison of the method using Stratified 10-Fold Cross Validation with the Area Under Curve (AUC) classification merit criterion shows that the results of NBC performance with SMOTE are better shown with an AUC value above 80% for testing data in each topic in each period. The study ended with visualization using the Lexical Salience-Valence Analysis (LSVA) Quadrant which is based on the calculation of word salience and word valence to see the level of positivity or negativity as well as observe changes and shifts in visitor experience preferences with the pandemic

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Bali, Latent Dirichlet Allocation, Lexical-Salience Valence Analysis, Naive Bayes Classifier, SMOTE, Tourist Accommodation, Akomodasi Wisata, Bali, Latent Dirichlet Allocation, Lexical-Salience Valence Analysis, Naive Bayes Classifier, SMOTE
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA278.55 Cluster analysis Q Science > QA Mathematics > QA279.5 Bayesian statistical decision theory. Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science) Q Science > QA Mathematics > QA76.F56 Data structures (Computer science) T Technology > T Technology (General) > T57.5 Data Processing
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Fadhila Salsabila
Date Deposited:	09 Aug 2024 05:25
Last Modified:	09 Aug 2024 05:25
URI:	http://repository.its.ac.id/id/eprint/114986

Actions (login required)

View Item