大学生心理健康发展偏差趋势预测算法

Prediction algorithm for deviation trend of college students' mental health development

摘要:

当大学生的心理压力过大时会出现严重的心理问题，需要通过有效的方式对其心理健康发展情况进行预测。基于此，本文对大学生心理健康发展偏差趋势预测算法进行研究。通过加权关联规则量化处理大学生心理数据，采用多边邻域矩阵表征心理数据发展趋势时序，基于自相关函数预测大学生心理发展趋势偏差，完成预测方法设计。实验结果表明：以不同的压力事件作为影响大学生心理健康发展的测试内容，分别通过平均绝对误差、均方误差、均方根误差进行预测对比，新方法能够实现最小误差下的发展趋势预测，误差值均在0.1以下。该方法能够保证对大学生心理健康发展的真实判断，具有应用价值。

Abstract:

When college students are under excessive psychological stress, they may experience severe psychological problems, necessitating effective methods to predict their mental health development. Based on this, the prediction algorithm is studied for the deviation trend of college students' mental health development. By quantifying college students' psychological data through weighted association rules and characterizing the development trend of psychological data using a multi-edge neighborhood matrix, the prediction of the deviation in the mental health development trend of college students is achieved based on the autocorrelation function, completing the design of the prediction method. Experimental results show that, using different stress events as test content affecting the mental health development of college students and comparing predictions through mean absolute error, mean squared error, and root mean squared error, the new method can achieve trend prediction with minimal error, with error values all below 0.1. This method can ensure an accurate judgment of the mental health development of college students and has practical application value.