Readability indicator based information retrieval method

Abstract

The present invention discloses a readability indicator based information retrieval method. The method comprises: in a searching process by using a search engine, sorting documents, which meet a search condition, according to a relevance between the documents and a query keyword; and organizing the documents that meet the search condition, a relevance sorting and a readability score into a page and returning the page to a user, wherein a text readability score equals to M*(N*average Chinese stroke number+(1-N)*difficult Chinese word frequency)+(1-M)*(P*average English character number+(1-P)*difficult English word frequency), M adjusts a weight proportion of Chinese and English readability, N adjusts weight proportions between an average Chinese stroke number indicator and a difficult Chinese word frequency indicator, and P adjusts a weight proportion between an average English character number indicator and a difficult English word frequency indicator. According to the method disclosed by the present invention, the readability score of the document is returned after retrieval, so that the user can conveniently and rapidly extract a relatively readable portion from the documents with a relatively high relevance, so that retrieval efficiency is improved.

Description

technical field [0001] The invention relates to an information retrieval method, in particular to an information retrieval method based on a readability index. Background technique [0002] Information retrieval refers to the activity of obtaining information resources related to information needs from a collection of information resources. In modern society, information retrieval has become an important way for people to discover and acquire knowledge and information. For traditional information retrieval, after the user submits a series of queries to the retrieval system, the retrieval system returns a list of results for the user to select and read according to the "correlation" between the document and the query and the "importance" of the hyperlink structure. The specific process Such as figure 1 As shown in the figure, the interactive process of traditional information retrieval is shown in the figure. When a user submits a query to a search engine, the search engine...

AI Technical Summary

Problems solved by technology

Recently, in order to improve the accuracy of readability calculation methods, some readability methods try to use the idea of ​​machine learning to transform the readability calculation problem into classification and prediction problems, such as using Support Vector Machine (SupportVectorMachine) [5] , Regression [6] , Interpolation Prediction (InterpolationPrediction) [7] etc. However, these methods are difficult to support unsupervised online prediction of readability

[0006] As of now, search engines do not display the ability to indicate the readability of search results relative to the user

Images