To advance human resources management and talent employment in higher education institutions, this paper proposes a theme-based capability perception person-job matching neural network (TAPJFNN) by incorporating textual theme features into the APJFNN model. Real recruitment data from R University is selected as experimental data to explore the application of big data technology in intelligent recruitment. The results show that the proposed TAPJFNN model can effectively model the matching degree between talent and job positions. Compared to the APJFNN and APJFNN models, the performance of the proposed model is superior. Additionally, in terms of AUC, TAPJFNN outperforms TAPJFNN_preliminary by approximately 3%. This clearly validates the effectiveness of the proposed model in university human resources management systems.
