Multiclass Intent Classification for Chatbot Based on Machine Learning Algorithm
Wan Mohd Amir Fazamin Wan Hamzah, Mohd Kamir Yusof, Ismahafezi Ismail, Mokhairi Makhtar, Hasnah Nawang, Azwa Abdul Aziz · 2022 Seventh International Conference on Informatics and Computing (ICIC) · 2022
In recent years, the use of Chatbots has grown significantly in various industries, including support systems, education, health care, tourism, entertainment, and banking. Chatbot for education can provide instant feedback in interactions session with students. The responses from Chatbot can be based on machine learning algorithms or multiple heuristics techniques to select responses from the predefined library. The generated response depends on the user's intent of using the Chatbot. There are various classes or categories in the user's intent. However, the user's intent class cannot simultaneously belong to multiple classes. Therefore, this research proposed a multiclass intent classification for the Chatbot based on the machine learning algorithm. The findings of this research showed that Linear SVC is the best machine learning algorithm model for multiclass intent classification. The results of the analysis proved the accuracy of the prediction using Linear SVC.