Heat-sensitive unstructured data retrieval ranking optimization algorithm

Abstract

The invention provides a heat-sensitive unstructured data retrieval ranking optimization algorithm. The retrieval ranking optimization algorithm is provided with: unstructured data object, which at least comprises a plurality of data attributes or attribute classes; The service consumer task model missonda [] at least comprises a plurality of task attributes; And the popularity score calculation module compare () calculates the popularity score of the retrieval result according to the matching degree of the unstructured data attribute and the task attribute of the service consumer. According to the retrieval ranking optimization algorithm, the heat score of the retrieval result is calculated according to the matching degree of the unstructured data attribute and the service consumer task attribute, the retrieval result is ranked based on the heat score, and therefore optimization of the keyword retrieval result is achieved, and the retrieval result better conforms to the user preference.

Description

technical field [0001] The present invention relates to information retrieval technology, in particular to a method of calculating the popularity score of retrieval results based on the matching degree of unstructured data attributes and service consumer task attributes, sorting retrieval results based on popularity scores, and optimizing keyword retrieval results , Unstructured data retrieval ranking optimization algorithm with high accuracy of retrieval results. Background technique [0002] With the continuous advancement of network and communication technology, today's era is already an era of highly developed information. On the one hand, people can quickly obtain massive information resources, which undoubtedly brings a lot of convenience to life; on the other hand, unprecedented information resources The huge size also brings certain troubles to people, often making people dazzled and difficult to accurately obtain the resources they need. Therefore, information retr...

AI Technical Summary

Problems solved by technology

[0003] In the existing data retrieval technology research, in order to be able to perform cloud search on distributed information resources and obtain various information resource characteristics and content, information resource services will use web crawlers, combined with data mining, artificial intelligence, natural language processing and other related Technology realizes automatic search and intelligent search of information resources. This type of method can realize massive search of unstructured data, but this method mainly collects unstructured data and sorts retrieval results based on the degree of overlap of keywords. Structured data attributes and service consumer task attributes cannot be accurately matched, making the accuracy of retrieval results unsatisfactory and unable to meet user preferences

Images