Research on recognition and intervention of behavior sequences in virtual museum learning

Abstract

Learning in virtual museum can transcend the limits of time and space. The virtual museum that combines expertise in different disciplines provides a virtual learning environment for college students, but how to intervene in museum learning has been unclear. Targeted at this question, this study selected 2030 majors in clinical medicine from a certain university and the final results exhibited four types of learners who are of high, medium, low and absent museum immersion, respectively. When the learners visited the virtual museum, their behavior data were collected backstage and later used as data source. The method of fuzzy c clustering analysis was utilized to test the behavior recognition results of virtual museum learning, and lag sequential analysis (LSA) was used to carry out sequential transformation of learning behaviors in virtual museum. In this study, the four types of learners were subsumed under two broad categories of middle & high museum immersion and low & absent museum immersion. The importance of behavior was identified with random forest algorithm, and the intervention mechanism of museum teaching was designed according to the analysis results. Specifically, such strategies as museum support, voice guidance, video guidance, sub-museum ordering, rewards points on the list, etc. were used to study the museum learners in need of intervention. The results showed that the learning state of some learners was significantly improved.