Conversational Artificial Intelligence in Natural Language Processing Application with Lifelong Learning

Abstract

Conversational AI bot known as chatbot has many roles in human life today, such as answering uncomplicated questions on E-commerce website pages or even help as an assistant like Siri and Google assistant. However, the chatbot is currently limited due to insufficient knowledge available on a predetermined training dataset. So when a chatbot receives an unfamiliar question, it may be hard to understand the question. Therefore, we formed a chatbot using the lifelong learning method. Thus, that chatbot can conduct training on unfamiliar incoming data without train the model from scratch. The performance benchmark the system used is the average confidence score of the tests carried out. The system has an average confidence score of 0.80952.

Tasks

Reproductions