Analysis of the Impact of AI on College Students’ Employment Pressure Based on the Random Forest Algorithm

为了能够更好地分析AI对大学生就业带来压力，本研究融合经典统计检验(T检验)与机器学习算法；通过贝叶斯优化对随机森林超参数调优；利用最佳参数组合对样本进行建模，挖掘出AI给大学生带来就业压力的主要影响因素。模型评价指标如下：准确率(0.8755)，精确率(0.8857)，召回率(0.9051)，F1分数(0.8953)，AUC-ROC值(0.9399)，AUC-PR值(0.9441)。结果表明，本研究的模型分类效果具有优越性。我们通过建立模型并分析得出AI的专业关联、AI行业与就业冲击接纳等指标因素对大学生的AI就业压力感知影响重大。我们的研究结果可以为高校应对AI时代的学科调整和大学生就业指导以及为大学生的自我提升与就业提供了相应的理论支持，帮助大学生更好地应对AI时代的就业挑战。In order to better analyze the pressure brought by AI on college students’ employment, this study integrates classical statistical tests (t-test) and machine learning algorithms. The hyperparameters of the random forest are tuned through Bayesian optimization. The optimal parameter combination is used to model the samples, and the main influencing factors of AI-induced employment pressure on college students are excavated. The model evaluation indicators are as follows: accuracy (0.8755), precision (0.8857), recall (0.8832), F1-score (0.8953), AUC-ROC value (0.9399), and AUC-PR value (0.9441). The results show that the classification effect of the model in this study is superior. Through model building and analysis, we have found that factors such as the correlation between majors and AI, and the acceptance of AI-related industry and employment impacts have a significant influence on college students’ perception of AI-related employment pressure. Our research results can provide theoretical support for colleges and universities to cope with the discipline adjustment and employment guidance of college students in the AI era, as well as for the self-improvement and employment of college students, and help college students better cope with the employment challenges in the AI era.