AI CHATBOT TECHNOLOGY FOR VERIFYING PRODUCT WARRANTY INFORMATION FROM IMAGES

Abstract

This research aimed to develop a chatbot system that employs artificial intelligence and image processing to verify product warranty information from images. The objective was to enhance the accuracy of information retrieval and solve organizational issues in small and medium enterprises, which often rely on searching for data from paper documents such as tax invoices. This manual process is time-consuming and prone to data errors. The developed system connects to a database via LINE official account and Google sheets API to extract serial numbers from QR codes detected in product images. It then automatically displays the purchase date and warranty expiration date to users through the chatbot. The research was conducted using a systematic learning process framework to design and develop a prototype system, followed by performance evaluation based on accuracy criteria. The sample group consisted of five sales employees. The experimental results showed that the system could verify warranty information with an average accuracy of 90%.