Year: 2024
Type: conference-paper
Abstract: This paper describes the implementation of a BERT-based travel quiz system, including data pre-processing, model training and evaluation. The CrossWOZ public dataset was used to implement a domain-specific quiz task, which was processed into quiz-pair form and fine-tuned using a pre-trained BERT model. The project performed a detailed performance evaluation of the model using BLEU and ROUGE scores to assess the responses it generated, and loss function images were plotted during training to demonstrate the effectiveness of the model training. The results show that the BERT-based Q&A system has high accuracy and stability in handling Chinese Q&A tasks.
Authors: Yudi Geng, Hongzhi Yu, Fucheng Wan
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Funder: Northwest Minzu University
OpenAlex ID: W4411089160