
SentiShop: A Multilingual Sentiment Analysis and Extractive Summarization of Lazada Product Reviews Using Transformer-Based Algorithms
SentiShop is a web-based application developed to automate sentiment analysis and summarize multilingual product reviews on Lazada Philippines. The system addresses the challenges of information overload, language diversity, and code-switching commonly found in online product reviews. It combined two core Natural Language Processing (NLP) techniques: a fine-tuned transformer-based sentiment classifier (XLM-RoBERTa) and a graph-based extractive summarization algorithm (TextRank).
Group Members
Mentor
Topics
AVP
Gallery






