Article identification method and device, and storage medium

Abstract

The embodiment of the invention discloses an article identification method and device, and a storage medium. The embodiment of the method comprises the following steps that: selecting a correspondingarticle as a sample to construct the sample set of an original article identification model; then, obtaining the sample title vector and the sample main body vector of the sample set, training the original article identification model according to the sample title vector and the sample main body vector to obtain a trained article identification model; obtaining the title vector and the main body vector of an article to be identified; according to the trained article identification model, identifying the article title vector and the main body vector to obtain similarity information between thetitle vector and the main body vector; and according to the similarity information, determining whether the article to be identified is a title attractor article or not. By use of the scheme, the title attractor article can be automatically identified through the article identification model, artificial identification is not required in an identification process, and the identification efficiencyof the title attractor article can be improved.

Description

technical field [0001] The present invention relates to the field of communication technology, in particular to an article identification method, device and storage medium. Background technique [0002] The content aggregator in the instant messaging application aggregates the articles sent by the instant messaging official account. When the content aggregation server corresponding to the content aggregator pushes articles to users, it will push corresponding articles to users according to the official account information subscribed by users, user interests, etc. [0003] At present, in order to attract readers, some article publishers (such as official account owners, article authors, etc.) will add some exaggerated, eye-catching, and false titles to the articles. The characteristics of these articles are that the titles are completely irrelevant or have little connection with the content of the article. , the title is false, deceptive, ambiguous, etc. Generally, such arti...

AI Technical Summary

Problems solved by technology

[0005] At present, fake news and headline party articles are mainly identified through manual review and user reporting. Usually, the main measure is to delete the article, which requires a lot of manpower investment, and the corresponding system and auxiliary tools need to be built. personnel training

Therefore, the current identification efficiency of headline party articles is relatively low and the cost is high

Images