Quantitative Evaluation and Path Optimization of College Students’ Employment Policy Based on Text Mining

Abstract

Based on the text mining method, the high-frequency word extraction of 20 policies to promote the employment of college students from 2015 to 2020 was carried out, and the key areas of concern of the policy texts were obtained, and the relationship between the policy priorities was visually displayed in the form of visual Knowledge Graph. On this basis, a PMC index model composed of 9 first-level indicators and 31 second-level indicators is constructed to quantitatively evaluate the employment policy of college students and select a representative policy P3 to draw the PMC surface map to trace the advantages and disadvantages of the secondary indicators of the policy and explore the path of policy optimization. The results show that the evaluation results of 19 of the 20 policies are acceptable or excellent, accounting for 95%, indicating that most of the current policies can effectively promote the employment of college students. However, representative policies can be optimized and upgraded according to two different ideas: “policy tendency-policy field-target-policy focus” and “policy tendency-policy field-policy effectiveness-policy focus-target-policy nature.”