Statement-type identification method and device, storage medium and electronic device

Abstract

The invention discloses a statement-type identification method and device, a storage medium and an electronic device. The statement-type identification method includes the steps that a statement identification request is obtained, and is used for determining whether to-be-identified statements issued by a target account number in target application belong to target types or not, wherein the targettypes are types of statements prohibited issuing in the target application; target characteristic vectors of the to-be-identified statements are obtained, and are used for showing the to-be-identified statements; according to trained classification models and the target characteristic vectors, type identification results are obtained, the target characteristic vectors are input of the trained classification models, the type identification results indicate whether the to-be-identified statements belong to the target types or not, the trained classification models are at least used for real-time identification of the target characteristic vectors through characteristic vectors of context statements, and the context statements are statements which are issued by the target account number in the target application and adjacent to the to-be-identified statements. By means of the statement-type identification method and device, the storage medium and the electronic device, the technical problem that the accuracy when whether user comments belong to harassing comments or not is detected with related technologies is low is solved.

Description

technical field [0001] The invention relates to the field of computers, in particular to a sentence type recognition method and device, a storage medium and an electronic device. Background technique [0002] In order to purify the network environment, in the client application, it is necessary to detect and combat the user's harassing speech. The related technology mainly realizes the detection of harassing speech by manually defining a rule dictionary and matching user speech with the rule dictionary for keywords. This method relies on manual dictionary updates, which consumes a lot of manpower, has a low detection rate for new words or unregistered words, and does not consider the context information of user speech. In other words, the accuracy of relevant technologies to detect whether user speech is harassment speech is relatively low. [0003] For the above problems, no effective solution has been proposed yet. Contents of the invention [0004] Embodiments of the...

AI Technical Summary

Problems solved by technology

[0004] Embodiments of the present invention provide a sentence type identification method and device, storage medium and electronic device, to at least solve the technical problem of low accuracy in detecting whether a user's speech is harassment speech in related technologies

Images