Make Chat-GPT 4 an Expert Reading Assistant

In this study, an end-to-end framework was constructed to consolidate key information into an intuitive, easy-to-read table or graphical interface by using ChatGPT4 to analyse news articles from the automotive industry and generate Python code. To explore ChatGPT-4’s potential as a professional working assistant, it was given the status of industry expert by using ‘expert tips’ to enable in-depth analysis and advice on article topics. Experiments were also conducted using the fetch_20newsgroups dataset to evaluate the model’s performance in this task. The experimental results show that ChatGPT-4 performs well as a professional working assistant and that the ‘expert prompting’ strategy successfully stimulates the potential of the large language model to answer as a leading expert. This study is potentially important for various professional domains.