Distributed semantic and sentence meaning characteristic fusion-based character relation extraction method

Abstract

The invention relates to a distributed semantic and sentence meaning characteristic fusion-based character relation extraction method, and belongs to the field of natural language processing. The method comprises the steps of firstly performing training in a small amount of marked corpora and a large amount of unmarked corpora by utilizing statistic word frequency features and a Bootstrapping algorithm to obtain a relational feature dictionary; secondly constructing a triple instance of a statement through an element distance optimization rule, and constructing a triple feature space by fusing distributed semantic information and semantic information; and finally performing true-false binary decision on a triple, and obtaining a character relation type by utilizing a confidence degree maximization rule. According to the method, automatic generation of the feature relation dictionary is realized; a conventional relational multi-class problem is converted into a triple true-false binary decision problem, so that a conventional machine learning classification algorithm is better adapted; and by utilizing the distributed semantic information, the accuracy of relational classification is improved.

Description

technical field [0001] The invention relates to a method for automatically extracting character relationships from Chinese texts or Chinese text sets, and belongs to the technical field of computer science and information extraction. Background technique [0002] Character relationship extraction is the accurate and rapid automatic extraction of character entities and the relationship between characters scattered in the text, which belongs to the research content of the field of information extraction. [0003] Information extraction technology (IE, Information Extraction) has to complete two major research tasks: entity recognition (EDR, Entity Detection and Recognition) and relational recognition (RDR, Relation Detection and Recognition). Among them, relationship recognition (also called "relationship extraction") is to extract existing relationships between entities from text, and the types of these relationships are predefined. Character relationship is a kind of entity...

AI Technical Summary

Problems solved by technology

[0019] Aiming at the problem of difficult feature selection of machine learning algorithms and insufficient feature analysis, resulting in poor classification results, the present invention proposes a character relationship extraction method that integrates distributed semantic and sentence features, which improves the accuracy of extraction from Chinese texts or Chinese text collections. The effect of automatically extracting the relationship between characters

Images