Extraction method of Internet political and diplomatic news events

Abstract

A method for extracting news events of Internet politics and diplomacy disclosed by the present invention aims to provide an extracting method capable of improving the accuracy of event identification. The word set defines event categories, constructs trigger vocabulary and event category templates containing trigger words and event argument roles for each type of event; combines text-dependent syntax to analyze, identify and extract event elements in foreign and political fields. The single-text preprocessing operation is completed, and the category event trigger words are calculated and expanded based on the sememe similarity; the sentence that meets the threshold of similarity is taken as the candidate event sentence. Filter the event elements that satisfy the event category template, and extract the entity elements in the event sentence; then fill the event elements into the corresponding argument roles according to the event template; filter the candidate event elements that meet the category template; generate the event structure according to the event template Description file to build an event library in the field of foreign affairs.

Description

technical field [0001] The invention relates to the technical field of text information extraction, in particular to a method for extracting Internet political and diplomatic news events. Background technique [0002] With the rapid development of science and technology, various sources of news data emerge in an endless stream, resulting in a rapid increase in the amount of multi-source, multi-category, and heterogeneous news data. As an important data source of open source intelligence, news data has the characteristics of high real-time and massive. How to find the desired target information from a large amount of unstructured news data, and how to conduct in-depth mining, analysis, and prediction of the attention target in a large amount of unstructured news data, is the data situation of various countries in the face of massive news data. Perception, risk warning and other key concerns and problems to be solved urgently. [0003] Structural transformation of unstructur...

AI Technical Summary

Problems solved by technology

The second is the filling of the content of the event argument role. In one sentence, multiple event elements may be extracted for the same event element type. How to select appropriate elements to fill the content of the event argument will also affect the accuracy of the final result of event extraction. profound influence

Existing technology extracts event trigger words based on the end-to-end event extraction model of deep neural network. The construction of the trigger word table is mainly by calculating word frequency and selecting relevant verb keywords as trigger words. Although deep learning can greatly reduce the feature engineering of manual participation To "fit" the training data, but this does not mean that people are not required to participate in the selection of features at all, especially the data is simply massive political and diplomatic news events, and the manual construction of trigger vocabulary by experts is a very heavy workload, and it is very It is cumbersome, unless it is clear which data has potential value, how to do proper preprocessing and how to transform and achieve what goals, otherwise deep learning cannot be applied in the field of political and diplomatic news

Images