Conversational Artificial Intelligence

13Evaluation of Chabot Text Classification Using Machine Learning

P. Kumaraguru Diderot^1*, K. Sakthidasan Sankaran¹, Malik Jawarneh², Hriakumar Pallathadka³, José Luis Arias-Gonzáles⁴ and Domenic T. Sanchez⁵

¹Department of ECE, Hindustan Institute of Technology and Science, Chennai, India

²Faculty of Computing Sciences, Gulf College, Al-Khuwair, Oman

³Manipur International University, Manipur, India

⁴University of British Columbia, Lima, Peru

⁵Cebu Technological University-NEC, Cebu City, Philippines

Abstract

A chatbot is a type of software with artificial intelligence (AI) that is meant to talk like a person, usually over the internet. More and more people use chatbots. Every chatbot is built from the ground up with the main idea that it can have an intelligent conversation with a human user (often through text messages) and respond to their questions in the right way. This chapter gives a detailed literature review of the different ways that chatbot text features can be optimized. In the literature review section, there is also a detailed look at how machine learning techniques can be used to classify texts and group them together. A framework for classifying chatbot text based on machine learning is also shown. The normalization method is used to remove noise during preprocessing. Particle Swarm Optimization is used to improve the features, and KNN, SVM, and Naïve Bayes techniques are used to classify the text.

Keywords: Chatbot text classification, preprocessing, normalization, ...