Event detection method for multi-hop neighbor information fusion based on graph structure

Abstract

The invention provides a multi-hop neighbor information fusion event detection method based on a graph structure. The method comprises the following implementation steps: (1) generating a training set; (2) constructing a multi-hop neighbor information fusion network based on a graph structure; (3) training a multi-hop neighbor information fusion network based on a graph structure; and (4) detecting events in the natural language text. According to the method, the multi-hop neighbor information fusion network based on the graph structure is constructed, the multi-hop syntactic information in the syntactic dependency tree is utilized, the multi-label attention mechanism is used for fusing the multi-hop syntactic information, syntactic features which are more effective for event detection are extracted in a targeted mode, and the event detection accuracy and efficiency are improved.

Description

technical field [0001] The invention belongs to the field of computer technology, and further relates to an event detection method based on multi-hop neighbor information fusion of graph structure in the field of natural language processing technology. The invention can detect the event category in the natural language text by detecting the trigger word that expresses the event in the natural language text. Background technique [0002] Event detection is an important task of information extraction in natural language processing. The main goal of this task is to identify the event instances presented in the text and determine the corresponding event type, which is widely used in intelligent transportation, social media, network public opinion analysis, event knowledge graph and other fields. Event detection aims to detect trigger words for events in text. The methods based on feature engineering rely on artificially designed features and lack scalability; the methods based...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to address the above-mentioned deficiencies in the prior art, and propose an event detection method based on multi-hop neighbor information fusion based on a graph structure, which is used to solve the efficiency of the event detection method purely based on sequence in the event detection of the prior art. Low problems, problems that may ignore partial dependencies based on the GCN structure, and problems caused by the lack of attention mechanism that is difficult to distinguish the importance of different parts of the features

Images