Event joint extraction method fusing local features and deep learning

Abstract

The invention discloses an event joint extraction method fusing local features and deep learning. The event joint extraction method fusing local features and deep learning comprises the steps of performing entity extraction, part-of-speech and dependency analysis and PV-utilization; training document vectors by using a PV-DM model learning hidden features, and entering the GCN; thirdly, identifying a memory unit between the event element and the event trigger, and helping to identify an event type and an event argument; and finally, marking the final event type by using the CRF, and identifying event elements by using a Softmax function in a full connection layer. The method has the beneficial effects that the subject information in the article is learned by means of the document vector, then the relation between events is mined by means of the graph convolutional network, and finally the information between the event type and the event element is learned by means of the memory unit and the local feature, so that the event joint extraction is completed, and the recognition performance is improved.

Description

technical field [0001] The invention relates to the field of event extraction, in particular to an event joint extraction method integrating local features and deep learning. Background technique [0002] How to quickly and accurately extract valuable information from the huge amount of information data on the Internet has become a major problem that people face. In this context, information extraction came into being. The tasks of information extraction include entity recognition and extraction, entity resolution, relation extraction and event extraction. Information is the objective fact that a specific person or thing interacts with a specific place at a specific time. Event extraction is to extract events of interest to users from unstructured information, and store them in a database in a structured way for users to view. [0003] The history of event extraction can be traced back to the late 1980s. Driven by MUC (Message Understanding Conference), ACE (Automatic Co...

AI Technical Summary

Problems solved by technology

[0039] Bidirectional LSTM-based event trigger word and event type recognition by Mihaylov et al. lacks document-level information; Ferguson et al.’s event element recognition based on structural features spends a lot of manpower and material resources and cannot mine hidden features between words when forming features. And there is an error propagation problem in its pipeline model; Nguyen et al.'s event joint extraction based on recurrent neural network cannot effectively identify all events when targeting multiple types of event sentences

Images