Multi-task learning algorithm design and implementation strategy for intelligent police academy management system

This paper describes the basic architecture of the police academy management system from two aspects: system design and functional module design. Based on the student behavioral data included in the system, the multi-task learning method is used to mine the potential cognitive state of students, such as the degree of mastery of knowledge points, to accurately provide assessment functions such as test question recommendation and withdrawal warning. The MTL network built in this paper chooses a hard parameter sharing approach to better assess the behavioral characteristics of police academy students. In this study, two experiments were conducted on some knowledge tracking tasks using multitask learning algorithms, namely, data distribution minimization and migration between schools, which verified the prominence and effectiveness of MTL algorithms in data distribution minimization and migration between schools. In this paper, we also experimentally compare the MTL algorithm with other traditional recommendation algorithms on three real datasets, and the experimental results show that the MTL algorithm’s accuracy on all the three datasets is improved by more than 24% compared with the traditional collaborative filtering recommendation algorithm (CF).