A method and a device for acquiring document information

Abstract

The invention relates to a document information extraction method and device based on sequence labeling and a learning model. The method comprises the following steps: training at least one sequence labeling algorithm model to obtain at least one offline sequence labeling algorithm model; Determining the accuracy of the annotation information in each of the offline sequence annotation algorithm models, and converting a to-be-processed document into a text document; Obtaining document structure format property information from the to-be-processed document; And inputting the text document and the structural format property information into the offline sequence labeling algorithm model to obtain labeling information corresponding to the document information in the document. According to the method, the key information of the document can be extracted by using the sequence labeling technology. And by using a multi-model fusion technology, different key information in the document can be extracted by using an optimal model. In addition, business rule reasoning and calculation are carried out on a typeface extraction result, and the application range is wider.

Description

technical field [0001] The present application relates to the field of data processing, and in particular, to a method and device for obtaining document information. Background technique [0002] Natural Language Processing (NLP) is simply a technology that enables computers to understand human language. There are many application directions of natural language processing, including text classification, text clustering, abstract extraction, sentiment analysis, text review and other applications. Machines can assist or even replace humans to do some text-related work to a certain extent. [0003] In daily work, document writing, review, review, and revision are relatively common, such as custom-made writing and revision of contract documents, preparation and review of bidding documents, extraction of insurance clauses, information extraction and analysis of securities announcements, etc. At present, NLP technology is still lacking in text writing work, and the effect of many...

AI Technical Summary

Problems solved by technology

Some documents are short, the key information is relatively concentrated, and the document format and content are relatively simple, the key information is relatively easy to find

For texts with long text content and various formats and content, it is time-consuming and laborious to find key information

For example, a bond prospectus usually has hundreds of pages, with many contents and a complex document structure, and key information is often scattered in different places in the document. It is time-consuming and laborious to find these key information

[0004] Some existing methods can perform a certain degree of information extraction, but most of the existing methods are based on traditional technologies such as keyword search, text matching, and regular expressions. , resulting in poor information extraction function, which directly affects the subsequent links

Images