Web环境下本体和实体驱动的企业竞争情报获取机制研究

项目名称： Web环境下本体和实体驱动的企业竞争情报获取机制研究

项目编号： No.70803001

项目类型： 青年科学基金项目

立项/批准年度： 2009

项目学科： 金属学与金属工艺

项目作者： 赵洁

作者单位： 安徽大学

项目金额： 17万元

中文摘要： 随着Web 技术的快速发展，如何从Web 上及时有效地获取企业竞争情报成为竞争情报研究中的热点问题。已有的方法局限在"网页驱动"的网页搜集和文本搜索上，还没有提出有效的Web企业竞争情报抽取方法。本课题从实体和关系抽取角度，对Web竞争情报的语义进行了深入分析，提出了基于实体的Web竞争情报表示模型和基于ER模型的概念建模方法。同时，深入研究了Web竞争情报抽取中的若干关键问题，着重探索了基于Web网站的竞争对手情报抽取、基于Web的商业关系抽取、Web竞争情报可信性与博弈模型分析、网页动态关系抽取以及网页时态文本信息抽取、排序与索引等问题，并建立了一个基于时空信息的新闻主题搜索引擎。本课题的研究为基于Web的竞争情报抽取及应用奠定了基础，为网络时代竞争情报理论与应用的发展提供了新的思路。 本项目共完成学术论文25篇。本项目的研究成果既具有理论创新，又有实际系统开发，项目成果具有广阔的应用前景和可以预期的经济效益。

中文关键词： 竞争情报；Web；信息抽取

英文摘要： With the rapid development of Web technologies, how to effectively acquire comptitive intelligence from the Web has been a hot research issue. Existed methods employ a Web-page-driven approach and focus on Web page collection and text-based searching techniques. As a consequence, there are no effective approaches for the extraction of Web-based competitive intelligence. In this project, we first analyzed the semantics of Web-based competitive intelligence based on an entity-and-relation-driven point of view, and propose a representation model as well as an ER-based conceptual model for Web-based competitive intelligence. Furthermore, we studied some key issues related with Web-based competitive intelligence extraction, such as extracting competitor intelligence from the Web, Web-based extraction of business relations, game-theory-based credibility analysis on Web-based competitive intelligence, extracting dynamic relations from Web pages, and temporal-textual information extration, ranking and indexing. We also built a topic-based news search engine which utilized the spatiotemporal information extracted from Web pages. The studies conducted in the project aimed at providing a foundation for Web-based competitive intelligence extraction and application, and also offering some new insights for the theories and applications of competitive intelligence in the Web age. Through the execution of the project, we have published 25 papers in peer-reviewed journals or conferences. The results of the project are both of theoretical values and practical merits.

英文关键词： Competitive Intelligence; Web; Information Extraction