Educational Conversational Solution Using Contact Center AI

Abstract

Deep Learning-based chatbot systems have seen increased adaption in the educational domain in recent years owing to increased sophistication in the Artificial Intelligence (AI) domain. However, most communication between students and educational institutions is still performed physically and causes significant administrative overhead, especially during admissions. The primary objective of this paper is to design a chatbot to assist international and local students at educational institutions. The produced chatbot, NUBOT, aims to be a Proof of Concept and an additional way for those students to obtain general information about the organization. The proposed framework has implemented a Hybrid model approach that combines Informational Retrieval-based Neural Networks and Generative-base Neural Networks. The attention base mechanism of transformers is utilized on top of the open-source Rasa platform to aid with BERT (Bidirectional Encoder Representation Transformer) and DIET (Dual Intent and Entity Transformer) Classifier. Evaluation of the hybrid model approach has done by getting an intent confusion matrix and performance metrics like Precision, Accuracy, and Recall & F1-Score, which come out to be 94.7%, 96.0%, 96.0%, and 95.1%, respectively, at an average mean response time of 316.43 ms per query on an average basis. From a performance viewpoint, the developed chatbot has been compared with a state-of-the-art chatbot and outperforms it. Thus, it concluded that NUBOT successfully fulfilled its goals and is highly scalable, able to handle wider scopes and vague inputs easily.