Research On Ranking Method Of The Result Documents Of Mathematical Expression Retrieval

Mathematical expression is an important part of many kinds of literature,especially the scientific literature.With the increasing number of digital document resources,obtaining the required documents by retrieving mathematical expressions has become an urgent problem to be solved.The sorting of the result documents obtained by the mathematical expression retrieval is the premise of improving the performance of the mathematical expression retrieval system and realizing the orderly output of the retrieval result documents to better meet the user's needs.In view of the complexity of mathematical expression structure,rich grammatical and semantic transformations,and the diversity of document attributes which makes it difficult to sort the result documents of mathematical expression retrieval,the thesis proposes a new method of sorting the result documents of mathematic expression retrieval system based on hesitation fuzzy set through using the advantages of hesitant fuzzy sets in dealing with multi-attribute and multi-affiliation decision problems.Firstly,the attributes and indicators affecting the ranking of retrieval results of the mathematical expressions document were analyzed,summarized and extracted;Then,the corresponding membership functions were constructed by analyzing evaluation attributes and indicators;Finally,the ranking of result documents in mathematical expression retrieval system was realized through obtaining the hesitant fuzzy similarity between the retrieval result document and the mathematical query expression by calculating hesitant fuzzy distance.In this thesis,19964 documents containing 345027 mathematical expressions are analyzed experimentally,and the sorting method of retrieval results based on hesitant fuzzy set is compared with the existing sorting method,which shows the effectiveness of the proposed method.