Wardoyo, Naufal Farras Dhafran (2024) Analisis Sentimen Berbasis Aspek Menggunakan Support Vector Machine Terhadap Ulasan Smartphone di Marketplace Indonesia. Undergraduate thesis, Universitas Muhammadiyah Malang.
PENDAHULUAN.pdf
Download (978kB) | Preview
BAB I.pdf
Restricted to Registered users only
Download (300kB) | Request a copy
BAB II.pdf
Restricted to Registered users only
Download (344kB) | Request a copy
BAB III.pdf
Restricted to Registered users only
Download (553kB) | Request a copy
BAB IV.pdf
Restricted to Registered users only
Download (1MB) | Request a copy
BAB V.pdf
Restricted to Registered users only
Download (186kB) | Request a copy
POSTER.pdf
Restricted to Registered users only
Download (1MB) | Request a copy
Abstract
User review data on marketplaces is an invaluable source of information regarding products and services. However, the large amount of review data makes it difficult for both potential buyers to make purchasing decisions and for business actors to analyze review data to improve the quality of the products they sell, so sentiment analysis is needed to obtain deeper and more detailed information from a large collection of review data. In this research, the author uses the ABSA (Aspect-Based Sentiment Analysis) technique to extract aspect information and its sentiments contained in a collection of review data on the Indonesian marketplace. The research data used focuses on the smartphone domain, namely the Xiaomi Redmi Note 10S. The implementation of the ABSA model consists of four stages, namely data restructuring, data preprocessing, aspect and opinion extraction, and model training for aspect categorization and sentiment classification. The classifier algorithm used in the aspect categorization and sentiment classification stages is Support Vector Machine (SVM). There are two data scenarios tested, namely data with extraction results (I) and data without extraction (II). Evaluation results using the confusion matrix show that the best model from scenario II produces an average accuracy score of 93.75%, precision of 85.58%, recall of 74.08%, and f-measure of 75.25%.
Item Type: | Thesis (Undergraduate) |
---|---|
Student ID: | 201810370311233 |
Keywords: | aspect-based sentiment analysis, support vector machine, smartphone reviews, marketplace |
Subjects: | T Technology > T Technology (General) |
Divisions: | Faculty of Engineering > Department of Informatics (55201) |
Depositing User: | 201810370311233 dnaufalfarras |
Date Deposited: | 26 Apr 2024 01:44 |
Last Modified: | 26 Apr 2024 01:44 |
URI: | https://eprints.umm.ac.id/id/eprint/5806 |