现代电子技术2017，Vol.40Issue(21)：91-93,97,4.DOI:10.16652/j.issn.1004-373x.2017.21.025
基于机器学习的大学生自杀风险预测与分析
Prediction and analysis of college student suicide risk based on machine learning
摘要
Abstract
The available prediction methods of college student suicide risk has large prediction error and low prediction ac-curacy,therefore,a college student suicide risk prediction and analysis method based on machine learning is put forward. The factors influencing the college student suicide risk are acquired. The sample entropy is combined to analyze the subsequence of the factors influencing the college student suicide risk to perform the complexity analysis,according to which,the factors are classified and superimposed. The cross-validation theory and phase-space reconstruction theory are used to establish the correla-tion coefficient matrix for the suicide factors of college student,and combined with the support vector machine theory to deter-mine the decision function of college student suicide risk prediction. According to the decision function,the prediction model of college student suicide risk is established. The experimental results show that the proposed method has high analysis and predic-tion accuracy for college student suicide risk,strong practicability,and can provide the important theory basis for the active de-velopment of the suicide prevention and crisis intervention.关键词
机器学习/大学生自杀风险/预测方法/结果分析Key words
machine learning/college student suicide risk/prediction method/result analysis分类
信息技术与安全科学引用本文复制引用
丁楠..基于机器学习的大学生自杀风险预测与分析[J].现代电子技术,2017,40(21):91-93,97,4.
