Method and device for extracting page information

Abstract

The invention provides a method and device for extracting page information. The method includes the steps that S1, web pages of a whole network are obtained; S2, the obtained web pages are analyzed to be a document object model tree one by one, and according to size, position and cascading style sheet information of the pages, visual blocking treatment is performed on the web pages to obtain visual blocks; S3, on the basis of semantic features, the visual blocks are labeled to obtain labeled blocks; S4, address information tree constructed in advance is used for analyzing texts in the labeled blocks sentence by sentence, and the labeled blocks including the address information are identified to address information blocks; S5, names and corresponding address information of interest points in the address information blocks are extracted; S6, the extracted names and address information of the interest points are correlated to obtain structured information. Compared with the prior art, the method and device can automatically mine objects which are various in structure change and huge in quantity in the whole network, save labor cost and improve the accuracy rate and the recall rate.

Description

【Technical field】 [0001] The invention relates to the technical field of Internet information processing, in particular to a method and device for extracting page information. 【Background technique】 [0002] With the continuous development of the Internet and information technology, the Internet has become the main source of people's daily access to information. Since web pages are increasing exponentially every day, in order to enable users to quickly and accurately obtain the information they are interested in, information extraction is usually performed on these massive page data first. The task of information extraction is to structure the information contained in the text, so that people can obtain the information they need like querying a database. For example, the method of information extraction can be used to extract the name, address, contact number and other contact information of the entity contained in the webpage, and obtain the data of geographical points of ...

AI Technical Summary

Problems solved by technology

Using this method not only consumes huge labor costs, but also requires the mining objects to have the same structural characteristics in the page, so it cannot be applied on a large scale due to the limitation of labor costs and page structure consistency.

For the situation where the number of mining objects is huge and the page structure changes, such as obtaining the geographic interest point data of entities in the entire network, the existing template-based structured information extraction methods cannot meet the requirements of extraction accuracy and information recall rate

Images