Text event extraction method in combination of sparse coding and structural perceptron

Abstract

The invention discloses a text event extraction method in combination of sparse coding and a structural perceptron. The method comprises following steps: 1) normatively labeling and creating text data according to ACE or RichERE as training samples; 2) taking entities extracted as candidate entities for event trigger words and event parameters and extracting text features; 3) further extracting text distributive word vector features and learning sparse coding features; 4) utilizing training samples and extracted text features, training a classifier of the structural perceptron while recognizing trigger words and parameters related to events in texts; 5) inputting the classifier of the structural perceptron through the step 1 as for new text data and extracting text event information. The text event extraction method in combination of sparse coding and the structural perceptron has following beneficial effects: sparse coding expressions for distributive word vector features based on a neural network are utilized for enhancing text features; on the other hand, a model of the structural perceptron is utilized for learning event trigger words and recognizing event participants. Therefore, the better event extraction effect is obtained.

Description

technical field [0001] The invention relates to event extraction, in particular to a text event extraction method combining sparse coding and a structure perceptron. Background technique [0002] An event is a thing that occurs or appears. The event involves the entities (people, objects, etc.) participating in the event or affected by the event, and also involves all aspects of space and time in the universe. It is very important to understand events and their descriptions in text data, and event extraction is often a key component of applications such as machine reading, news summarization, information retrieval, and knowledge base construction. [0003] Usually, the goal of event extraction task is to extract event-related trigger words and participants (persons or objects) in the text. Current state-of-the-art methods for event extraction generally include three steps: first, use pre-trained named entity recognition tools to extract entities such as people, institutions...

AI Technical Summary

Problems solved by technology

However, most of these features require manual intervention, and the extraction process takes a long time and is not universal

In recent years, neural network and deep learning technology have become research hotspots, and unsupervised distributed word vector extraction methods in the NLP field are becoming increasingly common. This distributed word vector learning method is simple and general, and does not require manual labeling of data, but its drawbacks It does not have the interpretability and flexibility of common sparse expression features, so some studies have converted this distributed word vector into a sparse expression form that is easy to use in traditional NLP problems

Images