论文部分内容阅读
将网页内容分析和网页链接分析结合起来,构建了一个基于LDA和领域本体的竞争情报采集系统。实验结果表明,该系统能防止主题漂移的发生,带来较好的主题收获率。
Combining webpage content analysis and webpage link analysis, a competitive intelligence gathering system based on LDA and domain ontology is constructed. Experimental results show that the system can prevent the occurrence of theme drift and bring about a good theme harvest rate.