混合语义相似度计算优化模糊查询的智能信息检索算法

针对搜索引擎领域存在的因查询条件模糊而严重影响信息检索性能的问题,提出了一种混合语义相似度优化模糊查询算法。首先,根据相似度计算概念提出了与其接近的新概念提取方法;然后,利用TF-IQF模型将链接划分成标记,并使用这些标记组成的集合表示出查询;最后,构建二分图识别出相关的查询,使用二分图计算出查询相似度。实验结果表明,相比点击文档、相关查询和反向查询三种算法,取得了更好的全查率、检索精度及F-度量。

Intelligent Information Retrieval Algorithm Based on Fuzzy Query Optimized by Hybrid Semantic Similarity Computing

A hybrid method of semantic similarity is proposed. First of all, a new concept of extraction method is proposed which is close to the past. Then, TF-IQF model is used to divide the links into tags and the query is collected. Finally, constructing the bipartite graph to identify the related query, and use it to calculate the similarity query. The experimental results show that, compared to the three algorithms, clicking document, related queries and reverse query, this algorithm obtains the better check all rate, retrieval precision and F-metric.