WhatsApp Chatbot Customer Service Using Natural Language Processing and Support Vector Machine

Abstract

People have switched professions to become small business entrepreneurs during the COVID-19 pandemic. There are so many UMKMs born in 2020 – 2021 with little experience in how to do business. Moreover, there is a shortage of manpower, where young UMKMs are only run by one person due to the continuity with the pandemic, so owners do not have the time or energy to help answer customer’s questions. One way to solve this problem is to use a Chatbot. Usually, a chatbot is used as a tool for humans to communicate by providing digital-based information, such as answering questions about the types of goods sold in a store. Chatbots are used because of their ability to interact with many individuals at once quickly and accurately. So that this research designs and develops a text classification system on chatbots using the Support Vector Machine. The purpose of developing this system is to simplify and speed up the proses of product information exchanges to customers. With an average response of less than 1 second and an average accuracy of 87.75% in classifying text, the chatbot is effective and efficient. The chatbot will classify the text and respond accordingly when given a message. So that the benefits of designing and developing this chatbot are convenience, speed, clarity in the process of customer and seller interaction. Keywords— Text Classification, Customer Service, Chatbot, Machine Learning, Support Vector Machine