Development of the Topic Tagging System for Thai and English-Translated Web Contents

This paper presents the creation and evaluation of an AI-driven web application for topic tagging, utilizing both Thai and English-translated data. It tests and compares various machine learning models, including random forest, k-nearest neighbor, and long short-term memory neural network, to ascertain their effectiveness in topic tagging. The study use data from prominent Thai news sources. Data were preprocessed, modeled, and evaluated based on standard metrics like accuracy, precision, recall and F1 scores. Random forest was the best model and chosen for the development. The paper also discusses the system's architecture and a user experience survey that evaluates the application's usability and performance.