Method, device and storage medium for automatic construction of event corpus based on dual mode

Abstract

The embodiments of the present invention relate to the field of data processing, and disclose a method, a device, and a storage medium for automatically constructing an event corpus based on a dual mode. The method for automatically constructing an event corpus based on a dual mode includes: acquiring a first theme event keyword input by a user; performing retrieval according to the first theme event keyword to obtain a first theme event corpus, and expanding the first theme event corpus to obtain a first theme event corpus. Second-theme event corpus; the third-theme event corpus is obtained according to the correlation between the second theme event corpus and the theme, and the third event corpus constitutes a corpus. By providing an automatic construction method of event corpus based on dual mode, it does not need experts to label news information related to topic events, thus improving the efficiency of constructing corpus and saving labor costs. Moreover, all relevant event corpora can be automatically collected, making the corpus more complete and more accurate.

Description

technical field [0001] Embodiments of the present invention relate to the field of data processing, and in particular, to a method, device and storage medium for automatically constructing an event corpus based on a dual mode. Background technique [0002] In recent years, with the rapid development of network technology, Internet data has become the main source for people to obtain information due to the advantages of rapid updating, wide range, and easy access. According to statistics, the vast majority of network data is stored in the form of text, recording a large number of news events, and these news events often revolve around a certain theme. In the era of big data, all news events related to a certain topic are filtered out from massive data, and a corpus of news topic events is constructed, which is helpful for the mining and analysis of news events. [0003] The inventor found that there are at least the following problems in the prior art: when constructing most...

AI Technical Summary

Problems solved by technology

[0003] The inventors found that there are at least the following problems in the prior art: when constructing most news topic event corpora, experts often need to manually label news information related to topic events, which is not only inefficient but also high in cost

And for news events, generally a theme event will have many related sub-theme events, it is difficult to collect all relevant event corpus during manual labeling, resulting in incomplete corpus, incomplete coverage, and low accuracy

Images