Method for extracting Chinese institutional unit name from text information

Abstract

The present invention discloses a method for extracting a Chinese institutional unit name from text information. The method comprises: loading to-be-analyzed text information; performing matching of a front part tagging rule on the to-be-analyzed text information, marking the position of a front part word, and extracting information in coincidence with the front part tagging rule; performing posterior boundary identification processing on the extracted information to extract alternative company name data; performing matching of the front part tagging rule on the alternative company name data, and performing decision processing to obtain an alternative company name; and performing search verification on the alternative company name, judging whether the verification is successful, and if the verification is successful, obtaining a Chinese institutional unit name. According to method for extracting the Chinese institutional unit name from the text information, a front part word and a posterior boundary word of the Chinese institutional unit name are matched and marked, and the Chinese institutional unit name is extracted by combination with network search verification, so that the amount of operation is relatively small, the extraction speed is high, and the extraction precision is greatly improved; and therefore, the method can be widely applied to weighing instrument industries.

Description

technical field [0001] The invention relates to the field of text information extraction and mining, in particular to a method for extracting names of Chinese institutions from text information. Background technique [0002] With the rapid development of the Internet and its technology, the information on the network is growing explosively, and a large amount of information is presented in front of people in the form of electronic documents. People urgently need some automated tools to help them quickly find the real Important information, so information extraction research came into being, and named entity recognition research is an important part of information extraction. Named entity recognition refers to the identification of entities with specific meaning in text, mainly including names of people, places, institutions, dates, etc. Among them, the name of the institution is a relatively important category, especially the name of the institution in Chinese. In this app...

AI Technical Summary

Problems solved by technology

This recognition method needs to manually label a large number of corpus for training. The corpus has a large scale of features, the cost is relatively high, and the accuracy rate is not very high.

Images