A Search Engine Ranking Method and System Based on Improved Spectral Clustering

Abstract

The invention discloses a search engine sorting method and system based on improved spectral clustering, wherein the method includes: calculating the density-based similarity between every two initial webpage text contents, and determining the optimal category number of spectral clustering ; Calculate the class center vector and class mean value of the spectral clustering, calculate the connection degree between the new webpage text content and the existing spectral clustering, judge whether to add the new webpage text content in the spectral clustering, and judge the whether the spectral clustering needs to be merged with other spectral clustering; receive the user's query vector, calculate the correlation between the query vector and the existing spectral clustering, establish an initial result set of webpage links, and The correlation degree and the PageRank value of each web page in the initial result set are normalized and weighted to calculate the average value, sorted and presented to the user according to the size of the correlation degree and the PageRank value of the normalized weighted average value.

Description

technical field [0001] The invention relates to a search engine ranking method, in particular to a search engine ranking method and system based on improved spectrum clustering. Background technique [0002] With the development of search engine technology, users have higher and higher requirements for search engines. Although the search engine can search thousands of results according to the user's query, in general, the user only browses the first 30 or even the first 10 results. If the top results in the search engine have nothing to do with the user's query, The user's query efficiency will be significantly reduced. [0003] Traditional sorting methods include sorting methods based on classification, sorting methods based on text retrieval and sorting methods based on link integration analysis. Among them, the sorting method based on classification is an artificial classification method that has been abandoned; the sorting method based on text retrieval that only focus...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the lack of a solution in the prior art that can overcome the efficiency problem and the result correlation problem at the same time, the purpose of the present invention is to propose a search engine sorting method and system based on improved spectral clustering, which can combine text content and Effectively combine text links, propose a search engine ranking method and system that combines content and links, and improve users' search efficiency

Images