Entity relationship recognition method and apparatus

Abstract

The present invention relates to an entity relationship recognition method and apparatus. The method comprises obtaining a statement sequence from a target text in a corpus, and performing named entity recognition and dependency grammar marker on the statement sequence to obtain a marked text sentence; matching and retrieving the marked text sentence on basis of an entity relationship seed to obtain a training example; replacing the entity relationship seed word in the training example with predetermined identification, processing the training example after replacement combined with the named entity recognition and the dependency grammar marker, and generating a candidate rule; fuzzifying the candidate rule to obtain fuzzy rules; determining whether the fuzzy rules comprise a new rule; and retrieving the corpus according to the fuzzy rules to obtain a seed set when the fuzzy rules comprise the new rule, and using the obtained seed set as an entity relationship recognition result. Manual participation can be effectively reduced, dependence on the calibrated corpus is reduced, a new entity relationship can be found timely, and the entity relationship recognition method and apparatus are self-adaptive to entity relationship mining in different fields.

Description

technical field [0001] The present invention relates to the technical field of information processing, in particular to an entity relationship recognition method and device based on grammatical analysis and self-expansion. Background technique [0002] Information extraction (Information extraction, abbreviated as IE) technology can help people quickly locate the information they really need in massive information. Information extraction is an unstructured natural language document as input to generate a fixed format and unambiguous format. data process. Information extraction is a research hotspot in natural language processing. There are two important directions in the information extraction system, one is to extract entities from the text, and the other is to determine the relationship between the texts. [0003] The main task of Named Entity Recognition (NER for short) is to identify and classify proper names such as names of people and places and meaningful phrases su...

AI Technical Summary

Problems solved by technology

Self-expanding technology first constructs entity-relationship schemas based on unstructured text data, and then uses these schemas to extract relationships from document collections. These new extracted relationships are used to generalize the original schema, and then a new round of extraction is performed. This cycle operates until the extracted relationship meets our requirements. Although this method has bootstrap learning ability, it cannot timely and effectively eliminate the wrong relationship patterns generated during the learning process, and if the wrong relationship patterns cannot be detected If it is discharged in time, it may introduce more wrong relations, and after further use, it will lead to the expansion of mistakes

Moreover, there is no method that combines syntax analysis and Bootstrapping technology to realize entity relationship recognition

Images