The usage of chatbots in education is growing by the day, due to the benefits it may provide to both learners and tutors. The goal of this work is to propose a chatbot that can assist tutors in achieving their educational aims while tutoring lessons, taking into account their individual pedagogical preferences and current emotions. The proposed chatbot is based on an intentional pedagogical process model, machine learning technologies and natural language processing. Besides, an emotion analysis module was introduced which will modify the chat-bot’s response based on the tutor’s emotion. Finally, after the implementation of the designed model, a use case is presented and an experimental campaign was conducted in order to demonstrate its efficiency. We have also compared our chatbot to current works to demonstrate the potential of the proposed technique. The implications of the findings were discussed shows promising results and suggestions were made to open the way to other future works.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

A Smart Educational Chatbot for Tutors’ Guidance

  • Ines Bayoudh Saadi,
  • Walid Bayounes,
  • Othmen Fares Ben Ayed

摘要

The usage of chatbots in education is growing by the day, due to the benefits it may provide to both learners and tutors. The goal of this work is to propose a chatbot that can assist tutors in achieving their educational aims while tutoring lessons, taking into account their individual pedagogical preferences and current emotions. The proposed chatbot is based on an intentional pedagogical process model, machine learning technologies and natural language processing. Besides, an emotion analysis module was introduced which will modify the chat-bot’s response based on the tutor’s emotion. Finally, after the implementation of the designed model, a use case is presented and an experimental campaign was conducted in order to demonstrate its efficiency. We have also compared our chatbot to current works to demonstrate the potential of the proposed technique. The implications of the findings were discussed shows promising results and suggestions were made to open the way to other future works.