简单介绍主题信息采集系统;从5个方面对其核心技术进行深入研究,包括种子页面生成、主题表示、相关度计算策略、爬行策略以及结束搜索策略等;详细讨论种子页面生成的人工方式、自动方式及混合方式,基于关键词的主题表示与基于Ontology的主题表示,多种相关度计算启发式策略比较,基本爬行策略与隧道技术以及结束爬行的多种情形等;在分析相关技术的算法、特点与应用情况的同时,针对主题信息采集特点提出相应的改进意见。 Briefly introduce the theme information collection system; From five aspects of its core technology in-depth study, including seed page generation, theme expression, correlation calculation strategy, crawling strategy and the end of the search strategy; detailed discussion of the seed page generation manual mode, automatic Methods and hybrid methods, topic-based keyword representation and Ontology-based thematic representation, a variety of correlation computing heuristic strategy comparison, the basic crawling strategy and tunneling technology and the end of the crawling of a variety of situations; in the analysis of related algorithms, Characteristics and application of the situation at the same time, the characteristics of thematic information collection put forward the corresponding improvements.