基于犹豫模糊集的数学检索结果排序研究

发布时间：2018-05-13 04:07

本文选题：数学表达式 + 犹豫模糊集合　；参考：《河北大学》2017年硕士论文

【摘要】：随着网络中文档数量的与日俱增,人们的检索需求也在不断增加,数学表达式是大量网络文档的重要组成部分,人们经常会希望通过数学表达式在搜索引擎中检索所需的相关文档,但目前这一需求还难以得到满足。针对数学表达式复杂的二维结构及语法特征造成的数学表达式及其相关文档排序困难的问题,分别提出面向数学表达式以及面向相关文档的数学检索结果排序方法,利用犹豫模糊集在处理多特征、多隶属度模式方面的优势,计算数学表达式间的相似度,构造文档得分函数,实现基于犹豫模糊集的数学表达式检索结果有序输出。首先,通过数学表达式解析算法对数学表达式特征进行提取分析,构造数学表达式相关评价属性集合,并对每个评价指标定义隶属度函数,通过犹豫模糊集合理论将每个数学表达式转换为对应的犹豫模糊集合计算相似度,并与数学检索系统相结合实现面向数学表达式的检索结果有序输出。在表达式相似度基础上,对网络中文档特征进行分析提取,构造文档评价属性及相应的隶属度函数,设计了基于犹豫模糊集的文档得分函数。在实验阶段对基于犹豫模糊集理论的数学检索结果排序方法进行了验证,并与现有排序方法进行对比研究,实验表明本文提出方法对数学表达式及相关文档的排序具有良好的有效性及适应性。
[Abstract]:With the increasing number of documents in the network, people's retrieval requirements are also increasing, mathematical expressions are an important part of a large number of network documents. People often hope to retrieve the relevant documents in search engines through mathematical expressions, but this demand is still difficult to meet. Aiming at the difficulty of sorting mathematical expressions and related documents caused by the complicated two-dimensional structure and grammatical features of mathematical expressions, a sorting method for mathematical retrieval results oriented to mathematical expressions and related documents is proposed, respectively. Taking advantage of the advantages of hesitating fuzzy sets in dealing with multiple features and multiple membership degree patterns, the similarity between mathematical expressions is calculated, the document score function is constructed, and the ordered output of retrieval results of mathematical expressions based on hesitating fuzzy sets is realized. Firstly, the feature of mathematical expression is extracted and analyzed by the analytic algorithm of mathematical expression, and the set of related evaluation attributes of mathematical expression is constructed, and the membership function is defined for each evaluation index. Based on the theory of hesitant fuzzy set, each mathematical expression is transformed into the corresponding fuzzy set to calculate the similarity, and combined with the mathematical retrieval system to realize the ordered output of the retrieval result oriented to the mathematical expression. On the basis of expression similarity, the paper analyzes and extracts the document features in the network, constructs the document evaluation attribute and the corresponding membership function, and designs the document score function based on the hesitant fuzzy set. In the experiment stage, the sorting method of mathematical retrieval results based on the theory of hesitant fuzzy sets is verified, and compared with the existing sorting methods. Experiments show that the proposed method has good validity and adaptability to the sorting of mathematical expressions and related documents.
【学位授予单位】：河北大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.3

【参考文献】