An Empirical Study on AI-Driven Text Mining for Graduate Thesis Quality Management in Higher Education

In recent years, artificial intelligence(AI) technology has developed rapidly and is widely used in various educational scenarios and administrative management in the field of higher education. This paper takes Text-Mining process of blind review comments of master's theses from a university as an example to explore the application effectiveness of AI technology in the quality management of master's thesis. Through empirical research, it provides a case reference for the application of AI in higher education management and a replicable integration program for similar institutions. This study used a three- phase approach to analyse blind review comments, namely independent use of AI (ChatGPT), independent use of traditional text analysis tools (NVIVO), and collaborative analysis combining the two. It was found that despite the significant advantages of AI in terms of processing speed and scale, the AI's output highly depends on the user's prompts and pre-training. Moreover, human intervention was critical to the stability and accuracy of AI's output.