Baoding university QA chatbot | |
| Author | Zhao, Hong |
| Call Number | AIT Thesis no.CS-24-01 |
| Subject(s) | Chatbots--Design and construction--Case studies User interfaces (Computer systems) Human-computer interaction |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Computer Science |
| Publisher | Asian Institute of Technology |
| Abstract | Baoding University faculty and staff currently obtain their own university information through their own websites or manually. However, it is hard to locate specific infor mation and time consuming. This study introduces a framework based on LangChain and Retrieval-Augmented Generation, aiming to build a question-answering(QA) chat bot for Baoding University. The framework mainly includes the following key parts: data preprocessing; query filter and classification; ensemble retriever; rerank and con text expand; and generative Chinese large language models. This study comprehen sively compares the performance of two query classification methods, two fine-tuning methods, five Bert retriever models, three retrievers, and four generative Chinese large language models in terms of accuracy, rouge score, context precision, context recall, answer relevancy, faithfulness, response time, and user satisfaction. Through a series of experiments, it was found that the combination of the BERT retriever and the fine-tuned qwen-1.8b-chat Chinese large language model, along with reranking and context expan sion for RAG, proved to be effective in the university QA chatbot. Furthermore, the QA results indicate that this framework can be applied to the construction of university QA chatbots, as well as to QA chatbots in other domains. |
| Year | 2024 |
| Type | Thesis |
| School | School of Engineering and Technology |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Computer Science (CS) |
| Chairperson(s) | Chaklam Silpasuwanchai |
| Examination Committee(s) | Chantri Polprasert;Attaphongse Taparugssanagorn |
| Scholarship Donor(s) | China Scholarship Council (CSC) |
| Degree | Thesis (M. Eng.) - Asian Institute of Technology, 2024 |