Research on Data Mining System of Reinforcement Learning in College Students' Career Path Planning

This study proposes a data mining system for college students' career path planning based on reinforcement learning, aiming to solve the shortcomings of traditional career planning methods in terms of personalization and dynamic adaptability. Traditional career planning methods often rely on static rules or manual settings, and cannot effectively cope with the rapidly changing job market and the differences in individual career needs. Therefore, this system uses data mining technology to extract key career characteristics from students' historical data, and realizes dynamic path planning through reinforcement learning algorithms to optimize career choices. To this end, this study designed an innovative algorithm that combines the advantages of dynamic programming and reinforcement learning, and defines a reasonable state space and reward function to cope with the uncertainty and complexity in career path selection. Experimental simulation results show that the proposed reinforcement learning algorithm is superior to traditional methods in multiple dimensions. Specifically, the accuracy of the planned path is improved by 15 %, which shows that the system can better formulate career development paths based on students' personal characteristics and market needs; the data mining efficiency is improved by <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{2 0} \%$</tex>, with significant advantages in feature extraction and processing time; in addition, the user satisfaction score is also significantly improved, reaching 4.7 (out of 5 points), showing that students highly recognize the path recommended by the system.