Named entity recognition model training method and device and information extraction method and device

Abstract

The invention provides a named entity recognition model training method and device and an information extraction method and device. According to the training texts in the set, entities in the semi-annotation information of the training texts are matched with entities in the corresponding training texts; based on a matching result, an effective named entity label of the corresponding training textis obtained; vector representation of each training text in the training text set is obtained; and based on the vector representation of each training text in the training text set and the effective named entity labeling, a named entity recognition model is trained based on deep learning to obtain a target named entity recognition model.

Description

technical field [0001] The following description relates to the field of information extraction, and more particularly, to a named entity recognition model training method and device, and an information extraction method and device. Background technique [0002] Now, information extraction is a relatively common problem faced by various industries. For example, in the industrial world, in the process of business personnel dealing with various business problems, various types of articles are important reference materials for business personnel. Business personnel need to dig out important information in articles every day, but the massive amount of articles is overwhelming. load. Taking the work of the Shenzhen Stock Exchange (“Shenzhen Stock Exchange”) as an example, a total of 265,985 announcements were disclosed in 2016, and 291,607 announcements were disclosed in 2017. With the increasing number of listed companies, this number will also increase year by year. The incre...

AI Technical Summary

Problems solved by technology

However, in the rule-based approach, since a lot of manual work is required to formulate the extracted expert rules, which is not fundamentally different from the costly and inefficient human querying of advertised entities

In addition, there are usually omissions in a large amount of manual work, so the formulated rules are not completely accurate. For example, once an error occurs in the formulation of a company's entity rules, it will lead to errors in entity recognition and cannot be realized in actual applications. An error occurred in the information extraction or information extraction

Images