基于基尼系数的决策树在涉恐情报分析中的应用

发布时间：2018-06-03 00:59

本文选题：基尼系数 + 反恐　；参考：《情报杂志》2017年04期

【摘要】：[目的/意义]利用数据挖掘技术对恐怖分子的各种日常信息,如购物、社交、交通、通话记录、视频等行为进行分析,对涉恐线索进行预警和排查,越来越成为国际反恐的通用手段之一。如何利用数据挖掘对大量的涉恐基础数据进行快速分类成为当前涉恐情报分析的研究热点。[方法/过程]将研究如何利用基于基尼系数的特征选择方法对涉恐人员的情报信息进行快速分类。分类步骤分为多个层次的属性分裂,其中每个层次中包括三个步骤,分别为计算样本集的基尼系数,计算不同属性的基尼系数,通过比较基尼系数选择分裂属性。[结果/结论]该方法可以对大量涉恐情报基础数据进行快速分类,提高反恐预警的效率。
[Abstract]:[objective / significance] using data mining technology to analyze the daily information of terrorists, such as shopping, social, traffic, phone records, video, etc., and to warn and search for terrorist clues. Increasingly, it has become one of the universal means of international terrorism. How to use data mining to quickly classify a large number of basic data has become a hot topic in the field of information analysis. Methods / processes will be studied how to use the Gini coefficient based feature selection method to quickly classify the intelligence information of persons involved in terrorism. The classification steps are divided into multiple levels of attribute splitting, in which each level consists of three steps, one is to calculate the Gini coefficient of the sample set and the other is to calculate the Gini coefficient of different attributes, and the other is to choose the split attribute by comparing the Gini coefficient. [results / conclusion] this method can quickly classify a large amount of terrorist intelligence data and improve the efficiency of anti-terrorism early warning.
【作者单位】：中国人民公安大学;湖南财经工业职业技术学院;
【基金】：中国人民公安大学基本科研业务费项目“大数据环境下反恐怖情报的数据挖掘分类方法研究”(编号:2015JKF01223) 国家社会科学基金项目“反恐维稳背景下边疆地区维稳战略研究”(编号:14BZZ028)的研究成果之一
【分类号】：C934;G350

【相似文献】