摘要
为有效提高罪犯教育改造质量,将基于关联规则的数据挖掘方法引入到再犯罪特征的分析之中,以某监狱罪犯数据库的刑释人员为样本数据,采用Johnson约简算法和Apriori关联算法对其进行属性约简处理和关联规则分析。结果表明:罪名、年龄、文化程度和刑期之间具有较强的正相关,即犯有盗窃罪前科、年龄小、文化程度低、刑期短是再犯罪的主要特征。该方法能揭示潜在的再犯罪规律,对刑罚机关具有参考价值,使其教育改造工作更具有针对性。
In order to improve the quality of education reform of criminals effectively, data mining based on association rules was utilized to explore the features of recidivism. Johnson reduction algorithm and Apriori algorithm were applied to analyze partial sample data of released prisoners in the criminal database of a prison. The analysis results show that the charge, age and cultural degree are positively related to the term of penalty. In other words, criminal record of larceny, younger age, low cultural degree and short term of penalty are main features of recidivism. The method of this article can reveal the potential recidivist. Therefore it has reference value for penalty organs and is one of useful tools to improve the quality of correction in prison.
作者
冯卓慧
冯前进
FENG Zhuohui FENG Oianjin(Department of Information Management, Zhejiang Police Vocational Academy, Hangzhou 310018, China)
出处
《浙江理工大学学报（社会科学版）》
2017年第1期57-60,共4页
Journal of Zhejiang Sci-Tech University：Social Sciences
基金
浙江省教育厅一般科研项目(Y201533854)
