微型电脑应用

针对高校学生就业去向预测这一问题无法快速获取精准预测结果的缺陷，提出了机器学习算法的高校学生就业去向预测方法。采集身份信息、专业成绩等高校学生就业去向预测相关数据，将所采集数据通过数据清洗、数据规约以及处理缺失值、异常值3部分完成数据预处理，利用特征选择算法依据完成预处理的数据获取最优特征子集，利用最优特征子集建立高校学生就业去向预测数据集，将预测数据集输入支持向量机分类器内实现高校学生就业去向预测。选取MATLAB软件作为仿真平台，仿真实验结果表明，该方法可有效预测高校学生就业去向，预测准确率与预测召回率均高于98%;预测时间开销低于200 ms。

Aiming at the problem in predicting the employment destination of college students, it is impossible to quickly obtain accurate prediction results, hence, the simulation of college student employment destination prediction based on machine learning algorithms is studied. We collect identity information, professional grades and other relevant data for predicting the employment destination of college students, complete data preprocessing through three parts, i.e., data cleaning, data specification, and processing of missing values and outliers, and use feature selection algorithms to obtain data based on the preprocessing. We use the optimal feature subset to establish a data set for predicting the employment destination of college students, and input the prediction data set into the support vector machine classifier to predict the employment destination of college students. MATLAB software is selected as the simulation platform. The simulation experiment results show that the method can effectively predict the employment destination of college students. The prediction accuracy and prediction recall rate are both higher than 98%; the prediction time cost is less than 200 ms.