摘要
腐败严重影响社会发展和民众信心,本文获取中纪委反腐通告、"天涯杂谈"新发帖和"百度热搜新闻词"三种不同来源的互联网文本语料从多个视角分析十八大以来的反腐成果.使用LDA话题模型探测不同语料中的反腐话题,并对语料的时空特性进行分析,通过官员履历构建"官员共职网络"探索腐败官员团伙并结合时间、级别和领域信息分析高级别官员的反腐策略.结果表明,腐败官员相关的不同语料的时空分布不同且涉及的腐败相关话题的重点不同,"官员共职网络"对于研究腐败官员的复杂关系具有重要意义.
The corruption greatly affects the development of the society as well as individual's confidence in the society. We analyze the fruits of anti-corruption from different perspectives based on three kinds of corpora: announcement on the website of Central Commission for Discipline Inspection, Tianya posts and Hot Search Words related News (HSWS). LDA topic model is used to detect anti-corruption related topics and the network of corruption officials is built. It is found that the temporal and spatial distributions of the different corpora related to corrupt officials are different. Anti-corruption related topics in the corpora have different emphasis. The corrupt officials' network where they co-work is of great significance for the study of their relationship.
作者
苑鹏佳
唐锡晋
YUAN Pengjia TANG Xijin(Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing 100190, China)
出处
《系统工程理论与实践》
EI
CSSCI
CSCD
北大核心
2017年第7期1833-1841,共9页
Systems Engineering-Theory & Practice
基金
国家重点研发计划基金项目(2016YFB1000902)
国家自然科学基金(61473284
71371107)~~