版权所有:内蒙古大学图书馆 技术提供:维普资讯• 智图
内蒙古自治区呼和浩特市赛罕区大学西街235号 邮编: 010021
作者机构:China Civil Affairs Univ Sch Social Work Beijing 102600 Peoples R China Hefei Univ Econ Sch Humanities & Law Hefei 230012 Anhui Peoples R China
出 版 物:《ALEXANDRIA ENGINEERING JOURNAL》 (Alexandria Engineering Journal)
年 卷 期:2025年第118卷
页 面:208-215页
核心收录:
学科分类:12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 08[工学]
主 题:Political practice Social work Feature selection Machine learning algorithms Data-driven decision making
摘 要:Social workers need political and policy skills to promote social justice and systemic change. To build these abilities, one must understand the factors that drive political and policy engagement. This study employed mutual information to identify key factors affecting political and policy practice among social work students, focusing analysis on the most important factors. After feature selection, a refined dataset was subjected to machine learning models, such as the standard support vector machine (SVM), ordinal logistic regression (OLR), and ordinal random forest (ORF). The results indicate that the SVM obtains the highest accuracy of 91.38%, with OLR and ORF following closely behind at 91.00% and 89.63%, respectively. These findings indicate that the standard SVM is particularly well-suited for this dataset because of its capacity to establish optimal decision boundaries for the classification of political and policy engagement levels. Despite its somewhat lower accuracy, the ORF is a competitive option, while the OLR captures ordinal connections well using interpretability. Predictive modeling can improve teaching methodologies and focus interventions to encourage political and policy practice among social work students, according to the study. This project provides data-driven insights to educators and policymakers to improve student political and policy engagement.