Analysis of Chatbot-Based Image Classification on Social Commerce LINE@ Platform

The rapid development of internet has influenced people's lifestyle to become more digital. Social commerce is one of the e-commerce categories where sellers offer their products through social media. LINE Messenger provides a place where sellers and buyers can communicate to carry out transaction processes. This paper discusses about the research on a chatbot that is useful for handling picture messages by providing product information replies. The chatbot was built using the Convolutional Neural Network method for the image classification process. The tests conducted found the chatbot to be able to reply to buyer messages with an accuracy level of 0.68.