With the progress of the times and the development of technology, the creation of artificial intelligence technology has brought a huge impact on the field of education. For the field of education, exploring curriculum relevance is one of its core elements. However, most of the traditional course relevance analysis is conducted from the perspective of course performance, ignoring the external factors that affect the performance such as test difficulty and student level. Since course text descriptions can describe a course more objectively than grades, it is necessary to apply text similarity analysis to course relevance research. The prevalent text similarity analysis is only achieved by word frequency statistics, ignoring the connection between textual contexts. In this context, this paper utilizes the Word Mover’s Distance (WMD) approach to analyze the relevance of course texts. The experimental results show the heat map of course relevance and demonstrate the superiority and effectiveness of the WMD method through the experimental comparison with the traditional text distance metric algorithm.