中文摘要：

网页蕴含了大量的企业竞争情报。然而，现有的企业竞争情报获取系统还缺乏直接从网页中获取竞争情报的能力。本文提出了一个基于网页实体关系抽取与融合的企业竞争情报获取系统框架。该系统通过对网页内容的抽取与融合，最终形成可信的企业竞争情报数据。论文首先讨论面向Web的企业竞争情报自动获取系统的总体结构，并重点阐述了其中的企业竞争情报获取方法、企业竞争情报融合机制等问题及其解决方案。本文的工作为进一步建立实用的Web竞争情报获取与融合系统奠定了基础。

英文摘要：

Web pages contain a large amount of enterprise competitive intelligence. However, current competitive intelligence systems are short of the ability to directly get competitive intelligence from Web pages. In this paper, we present a system to acquire enterprise competitive intelligence, which is based on the entity relationship extraction and fusion of Web pages. It uses the extraction and fusion techniques to process Web content, and finally forms trustable competitive intelligence for enterprises. The architecture of the system is first introduced, and some critical issues as well as the basic solutions are discussed, such as the acquiring methods for enterprise competitive intelligence and the filtering mechanism of competitive intelligence. Our work is expected to provide a fundamental framework for future implementation of a practical Web-based acquiring and fusion system of competitive intelligence.