基于意向决策支持的结构化小文本电力搜索引擎技术研究

Research on structured small text power search engine technology based on intention decision support

电力结构化小文本搜索过程中,由于知识库的差异,使得用户搜索查询词修改率较高,无法满足用户搜索需求。为此,提出了基于意向决策支持的结构化小文本电力搜索引擎技术。根据Web页面的链接关系,通过该URL与主题的相关度计算结果,完成采集数据的处理。根据意向决策支持技术,明确用户意向决策支持结果。构建情景设定知识库、情景分析知识库,并构建情景设定知识库、情景分析知识库,作为电力搜索引擎的运行基础。根据查询目标向量和电力搜索记录,计算相似度两者相似度。利用欧式语义距离计算电力搜索目标向量相似度,将相似度最高的查询结果作为结构化小文本搜索引擎算法的输出结果。实验结果表明:所提出的电力搜索引擎技术与常规技术相比,平均查询词修改率降低了23%、19%。

In the process of power structured small text search,due to the difference of the knowledge base,the user's search query word modification rate is high,which cannot meet the user's search demand.To this end,a structured small text power search engine technology based on intention decision support is proposed.According to the link relationship of the Web page,the processing of the collected data is completed through the calculation result of the correlation between the URL and the topic.According to the intention decision support technology,clarify the user's intention decision support result.Build a knowledge base of scenario setting and analysis,and build a knowledge base of scenario setting and analysis,as the operating basis of the power search engine.According to the query target vector and the power search record,the similarity between the two is calculated.Euclidean semantic distance is used to calculate the similarity of the power search target vector,and the query result with the highest similarity is used as the output result of the structured small text search engine algorithm.The ex-perimental results show that the proposed power search engine technology reduces the average query word modification rate by 23%and 19%compared with the conventional technology.