作 者:林平[1] 武海燕[2] 孙菲菲[1] 邱济 Lin Ping;Wu Haiyan;Sun Feifei;Qiu Ji(1. School of Counter Terrorism, People's Public Security University of China, Beijing 100038) (2. Department of Police Technology, Railway Police College, Zhengzhou Henan 450053) 机构地区:[1]中国人民公安大学反恐怖学院,北京100038 [2]铁道警察学院公安技术系,河南郑州450053 出 处:《情报探究》2017年第6期17-22,共6页Information Research 基 金:河南省高等学校重点科研项目计划资助项目“基于力导向算法的公安信息可视化略论探讨”(项目编号:15A520172); 河南省教育厅人文社会科学探讨项目“面向网络涉警舆情的略论探讨”(项目编号:2017-QN-086)阶段性成果 摘 要:[目的/意义]旨在帮助情报人员更快识别和挖掘涉恐文本。[措施/过程]基于情报工作视角,将阿拉伯语作为探讨对象,论述标准阿拉伯语文本处理流程,构建符合反恐业务需求的文本分类模型,并用模拟实验验证其可靠性,提出一个面向阿拉伯语的涉恐文本研判模型。[结果/结论]该模型符合实战运用需求。[Purpose/significance] The paper is to help intelligence personnel identify and mine text involving terrorism more quickly.[Method/process] The paper bases on the angle of intelligence work, and takes Arabic as research object, to discuss text processing flow of the standard Arabic, builds text categorization model for counter-terrorism's need, uses simulation test to verify its reliability, so that proposes an Arabic-oriented evaluation model for text involving terrorism. [Result/conclusion] The model meets the needs of practical application. 关 键 词:阿拉伯语 自然语言处理 反恐怖主义 文本挖掘 随机森林 ,阿拉伯语专业论文,阿拉伯语论文范文 |