A basic study of a method for clustering the gaze regions of a program on the basis of eye movement information during Program Reading

The objective of our research is to construct a system that is able to estimate the program comprehension ability of individuals engaged in programming learning, and to provide them with appropriate advice and assistance. To estimate the program comprehension ability of learners, we focus on the eye movement of learners when reading programs. In our approach, we divide the program into several regions, which are treated as nodes in a directed graph. This allows us to express how the learner read the program in what order and how much time was spent on each part. This is done using a weighted adjacency matrix. Subsequently, a clustering process is applied to the adjacency matrix in order to detect clusters of eye movements and extract the clusters in the program that the learner focused on. This paper presents the results of experiments conducted based on the approach described above, as well as the potential for future development of this research.