KMITL Expo 2026 LogoKMITL 66th Anniversary Logo

Aspect-Based Sentiment Analysis for E-Commerce Product Reviews

Abstract

In today’s rapidly expanding e-commerce environment, the massive volume of product reviews makes it crucial to summarize user opinions in a way that is both comprehensible and practically applicable. This research presents a system for analyzing product reviews using Aspect-Based Sentiment Analysis (ABSA), a Natural Language Processing (NLP) technique that identifies key aspects of a review (such as shipping, product quality, and packaging) and evaluates the sentiment (positive, negative, or neutral) associated with each aspect, allowing both consumers and merchants to gain more efficient access to in-depth insights. This project focuses on developing AI for Thai-language ABSA by utilizing WangchanBERTa, a model trained on Thai data, and comparing it with various standard approaches such as TF-IDF + Logistic Regression, Word2Vec + BiLSTM, and Multilingual BERT (mBERT/XLM-R) to assess their performance in terms of accuracy, speed, and resource usage. Additionally, a dashboard visualization is provided to help users quickly grasp review trends. The expected outcome is to create an AI tool that can be practically employed in the e-commerce industry, enabling consumers to make easier purchasing decisions and assisting merchants in effectively improving their products and services.

Objective

ปัจจุบันการซื้อขายสินค้าออนไลน์เติบโตขึ้นอย่างรวดเร็ว ทำให้ข้อมูลรีวิวสินค้าจากผู้บริโภคมีปริมาณเพิ่มขึ้นเป็นจำนวนมาก รีวิวเหล่านี้มีบทบาทสำคัญในการตัดสินใจซื้อของลูกค้าและการปรับปรุงคุณภาพสินค้าของร้านค้า อย่างไรก็ตาม ปริมาณข้อมูลที่มากเกินไปและความหลากหลายของรูปแบบการแสดงความคิดเห็นทำให้การสรุปและวิเคราะห์ข้อมูลเหล่านี้เป็นไปได้ยาก ผู้บริโภคต้องใช้เวลามากในการอ่านรีวิวจำนวนมากเพื่อสรุปแนวโน้มความคิดเห็น ในขณะที่ร้านค้าประสบปัญหาในการวิเคราะห์ข้อมูลรีวิวเพื่อปรับปรุงผลิตภัณฑ์และบริการ เพื่อแก้ไขปัญหานี้ งานวิจัยนี้นำเสนอการประยุกต์ใช้ Aspect-Based Sentiment Analysis (ABSA) ซึ่งเป็นเทคนิคใน Natural Language Processing (NLP) ที่สามารถแยกแยะ แง่มุมสำคัญของรีวิวสินค้า (Aspects) และวิเคราะห์ อารมณ์ของแต่ละแง่มุม (Sentiments) โดยอัตโนมัติ การนำเทคนิคนี้มาใช้จะช่วยให้ผู้บริโภคสามารถรับข้อมูลเชิงลึกจากรีวิวได้ง่ายขึ้น และช่วยให้ร้านค้าสามารถใช้ข้อมูลรีวิวเพื่อปรับปรุงสินค้าและบริการอย่างมีประสิทธิภาพ โครงงานนี้ยังมุ่งเน้นการศึกษาว่า แนวทาง AI แบบใดมีประสิทธิภาพสูงสุดในการทำ ABSA สำหรับภาษาไทย โดยเปรียบเทียบ วิธีการประมวลผลภาษาธรรมชาติแบบดั้งเดิม กับเทคนิคการเรียนรู้เชิงลึกที่ทันสมัย เพื่อให้ได้แนวทางที่เหมาะสมที่สุด พร้อมทั้งพัฒนา Dashboard Visualization ที่ช่วยให้ข้อมูลรีวิวถูกนำเสนอในรูปแบบที่เข้าใจง่ายและสามารถนำไปใช้งานจริงในอุตสาหกรรมอีคอมเมิร์ซ

Other Innovations

Design of Feature "BAAC Good Worker" in Line OA BAAC Family

คณะเทคโนโลยีการเกษตร

Design of Feature "BAAC Good Worker" in Line OA BAAC Family

Currently, Thailand is facing the issue of an aging agricultural workforce and a shortage of labor. The hiring process for agricultural workers is primarily based on word-of-mouth within a limited, narrow area. This may have indirect negative effects in various aspects, such as the lack of basic selection criteria (experience, skills), budget control, and hiring processes. As a result, the idea of creating a job platform specifically for farmers was conceived.Currently, Thailand is facing the issue of an aging agricultural workforce and a shortage of labor. The hiring process for agricultural workers is primarily based on word-of-mouth within a limited, narrow area. This may have indirect negative effects in various aspects, such as the lack of basic selection criteria (experience, skills), budget control, and hiring processes. As a result, the idea of creating a job platform specifically for farmers was conceived.

Read more
Analysis of Factors Affecting Productivity Improvement in Sugarcane Processing Plants.

คณะวิทยาศาสตร์

Analysis of Factors Affecting Productivity Improvement in Sugarcane Processing Plants.

Sugar production from sugarcane is a complex process that requires precise control. One of the major issues is sugar loss, which can result from various factors, particularly "burnt cane," before being sent to the mill. This affects the quality of the sugarcane and the efficiency of sugar extraction, along with the performance of the machinery and the properties of the cane, which impact the amount of sugar extracted. This study aims to analyze the factors that influence sugar loss in the sugar production process, using quantitative data from a sugar factory. Nine variables were examined, including mechanical efficiency, machine downtime per day, cane waiting time per day, sand content in cane juice, pol extraction efficiency, overall working time efficiency, cane juice purity, cane sugar content (C.C.S.), and burnt cane. The data were analyzed using correlation analysis to examine relationships between variables and regression modeling to predict sugar loss. The results showed that mechanical efficiency, cane sugar content, and the amount of sand or impurities in the cane juice were significantly correlated with sugar loss. Mechanical efficiency had a direct relationship with the amount of cane milled, which improved sugar production. On the other hand, burnt cane, or cane that was burnt before harvesting, resulted in reduced sugar extraction and impacted the quality of the sugar. Therefore, reducing sugar loss in the production process can be achieved by improving machine efficiency, reducing impurities in cane juice, and managing burnt cane, which will improve sugar production efficiency in the future.

Read more
DEVELOPMENT OF A WEBSITE FOR COLLECTING INFORMATION ON FARMERS YOUNG SMART FARMER CHANTHABURI PROVINCE

คณะเทคโนโลยีการเกษตร

DEVELOPMENT OF A WEBSITE FOR COLLECTING INFORMATION ON FARMERS YOUNG SMART FARMER CHANTHABURI PROVINCE

This study aimed to develop a website for collecting and organizing data on Young Smart Farmers in Chanthaburi Province. Data were collected through structured interviews with a sample of 30 participants. The information obtained was categorized and utilized to develop the website, which was subsequently disseminated to farmers and other stakeholders. The study also assessed user satisfaction with the website through a questionnaire, with data analyzed using descriptive statistics, including frequency, percentage, mean, and standard deviation.The results indicated that the sample comprised an equal proportion of male and female participants, with the majority (50.00%) aged between 36 and 40 years. Most respondents were Young Smart Farmers from the districts of Khlung, Laem Sing, and Kaeng Hang Maeo, each representing 13.33% of the sample. The majority of participants had attained a bachelor’s degree or equivalent (60.00%) and were primarily engaged in agricultural occupations (73.33%). The findings on user satisfaction with the website revealed a high level of satisfaction across all dimensions, ranked as follows 1) Website usability (Mean 4.97), 2) Overall satisfaction (Mean 4.93), 3) Content quality (Mean 4.91), 4) Practical benefits and applicability (Mean 4.87), and 5) Design and layout (Mean 4.85).

Read more