A Chinese named entity recognition method and system

Abstract

The invention discloses a Chinese named entity recognition method and system. The method comprises the following steps: S1, performing entity recognition based on rule matching on the target text to obtain a first named entity set; S2, using a statistical algorithm to perform entity recognition on the target text Identifying, obtaining a second named entity set; S3, after cleaning the first named entity set and the second named entity set, obtaining a recognition result. The present invention performs entity recognition on the target text based on rule matching and statistical algorithms respectively, and after cleaning the recognition results of the two, obtains the final Chinese entity recognition result, which can greatly improve the accuracy of Chinese entity recognition while ensuring the accuracy of Chinese entity recognition. The recall rate of the Chinese entity recognition is high, and the automatic recognition of the Chinese entity is carried out by the method, and the recognition speed is fast, which can be widely used in the field of text information processing.

Description

technical field [0001] The invention relates to the field of computer application and information processing, in particular to a Chinese named entity recognition method and system. Background technique [0002] Named entity is the basic information element in the target text, and it is the basis for correct understanding of the target text. Chinese entity naming recognition is an important basic tool in information extraction, syntactic analysis, machine learning and other application fields, and plays an important role in the process of natural language processing technology becoming practical. Chinese named entity recognition is to determine whether a string represents a named entity. In information extraction research, Chinese named entity recognition is currently the most practical technology. The commonly used method is purely based on hidden Markov, maximum entropy model recognition method. [0003] At present, due to the lack of regularity in the naming of Chinese ...

AI Technical Summary

Problems solved by technology

[0003] At present, due to the lack of regularity in the naming of Chinese company names, they are used more casually and often appear in the form of abbreviations, such as "Bank of China Co., Ltd." often appears in the form of abbreviations, such as "Bank of China" or "Bank of China", This brings difficulties to the identification and application of company names

In general, there are the following difficulties in identifying Chinese named entities such as Chinese company abbreviations: 1. In different fields and scenarios, the extension of the abbreviation is different

2. Certain types of entity names change frequently, and there are no strict rules to follow

4. The number is huge, cannot be enumerated, and it is difficult to include them all in the dictionary

In general, in the processing of Chinese target text, the effect of Chinese word segmentation greatly affects the recognition effect of Chinese named entities, which in turn affects the analysis and processing effect of the target text, resulting in low recall rate and slow recognition speed.

Images