A Novel Semi-Automated Chatbot Model: Providing Consistent Response of Students’ Email in Higher Education based on Case-Based Reasoning and Latent Semantic Analysis

Authors : Selim Buyrukoğlu, Yıldıran Yilmaz

Pages : 6-12

View : 22 | Download : 12

Publication Date : 2021-07-31

Article Type : Research

Abstract :Learning is a process that requires interaction, asking and responding to the question is one of the most crucial interactions in learning. Students ask questions to instructors through online learning systems or emails outside the class. Providing a consistent, correct, detailed and personalised response to these questions is essential for developing their skills and abilities about the subject. Several automated chatbot models have been developed to answer students’ questions consistently. However, these automated chatbot models lack the ability to comprehensively and reliably respond to students’ questions. Thus, other forms of Chabot models are required to provide detailed and reliable responses along with consistent responses. This study aims to develop a semi-automated chatbot model to answer students’ questions via email. This study also focuses on ensuring the consistency of answers. Therefore, a novel chatbot model is developed based on the semi-automated approach which supports new ways of answering students’ questions via email. A feasibility study was carried out to investigate and demonstrate the efficiency of the proposed model. The Case-Based Reasoning (CBR) cycle is adopted in the answering process to ensure the consistency of the answer. Semantic analysis tool was used to measure the semantic textual similarity between questions/sentences. The feasibility study results confirm that the proposed semi-automated assessment approach is feasible for use in higher education. Also, these results highlighted that this model enables the instructor to provide a consistent and personalised answer to students while considerably reducing the instructor’s workload.
Keywords : Chabot, Case-Based Reasoning, Latent Semantic Analysis, Consistent Response, Higher Education