A Hybrid Chatbot Model for Students Advising using Deep Learning Approach
Project Description

The objective of the proposed students advising chatbot is to eliminate the need of student advisor involvement as much as possible. It is a hybrid chatbot, which consists of both retrieval-based and generative models. In a situation where the retrieval-based model is applied, the proposed chatbot can answer questions automatically with trained answers and/or predefined answers stored in the repository. In another situation where the generative model is applied, the chatbot is capable of answering questions beyond the set of expected questions, or questions that demand for further student advising by student advisors.

Supervisor
KWOK James Sai Ho
Quota
2
Course type
UROP1100
Applicant's Roles

- Literature review
- System architecture
- System design and development for testing the proposed chatbot

Applicant's Learning Objectives

- Conduct literature review
- Approach research problems in a logistical way
- Conduct experiments and write up research papers/documents

Complexity of the project
Moderate