科技情报智能检索与语义分析  

Intelligent Retrieval and Semantic Analysis on Science and Technology Information

作　　者：袁林[1] 葛唯益 陈晓琳 YUAN Lin;GE Weiyi;CHEN Xiaolin(The 28th Research Institute of China Electronics Technology Group Corporation, Nanjing 210007,China)

机构地区：[1]中国电子科技集团公司第二十八研究所

出　　处：《指挥信息系统与技术》2019年第5期34-39,共6页Command Information System and Technology

基　　金：国家重点研发计划(2018YFC0806902)资助项目

摘　　要：为实现互联网上科技文字信息的有效获取和快速分析研读,研究了基于Heritrix增量式爬虫的科技情报搜集、基于科技情报语料库的统计机器翻译、面向主题-内容-结构的多特征文本自动摘要及基于支持向量机(SVM)向量空间模型的科技情报自动分类4项关键技术,设计了互联网科技情报搜集分析原型软件的技术架构和功能模块.试验表明,该原型软件能够有效提高科技情报信息的智能处理分析能力.To realize the effective acquisition and rapid analysis on the science and technology in-formation from Internet, four key technologies including the information collection based on Heritrix incremental crawler, the statistical machine translation based on the corpus of information, the automatic summary of multi-feature texts oriented to the subject, the content and the structure, and automatic classification of information based on the vector space model, support vector machine (SVM), are studied. The technical architecture and functional modules of the gathering and analysis prototype software for Internet science and technology information are designed. Experiments show that the prototype software can effectively enhance the intelligent processing and analysis capabilities of science and technology information.

关 键 词：科技情报 网页爬虫 文本摘要 文本分类 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]