Text content tag labeling method and device

Abstract

The invention relates to the technical field of television networks, in particular to a text content tag labeling method and device. The method comprises the steps that multiple feature tags of a classified text content matched with a target text content are acquired; the similarities between all feature gags and a word segmentation fragment of the target text content are calculated; the target text content is labeled with the feature tag of which the similarity is larger than a preset similarity threshold value. According to the text content tag labeling method, content sorting automation can be achieved, a many-to-many relationship between the content and the tags is achieved, various classification and aggregation requirements are met, iterative updating (self-optimizing) can be conducted in the sorting process, therefore, the method quickly adapts to sorting of the hot and popular content, and a user can more quickly and more precisely search for the content that the user wants.

Description

technical field [0001] The invention relates to the technical field of television networks, in particular to a text content label marking method and device. Background technique [0002] The CMS (Content Management System, content management system) deployed in the radio and television network environment is developed using the J2EE technology system, which can realize the classification and distribution of content, so that users can quickly search or retrieve the content they want. User's search is facilitated. [0003] However, with the development of network applications and the continuous expansion of product operation requirements, the existing content management system cannot quickly follow up a large amount of derivative content information, and often requires professionals to manually classify a large number of different new contents. Before classification, It usually requires professionals to visit the website in person, create a column, then select the appropriate...

AI Technical Summary

Problems solved by technology

[0003] However, with the development of network applications and the continuous expansion of product operation requirements, the existing content management system cannot quickly follow up a large amount of derivative content information, and often requires professionals to manually classify a large number of different new contents. Before classification, It usually requires professionals to visit the website in person, create a column, then select the appropriate text content to put in the corresponding column, and finally check the text content and quality of the website to ensure that the text content is included in the appropriate category. Through this completely manual It takes a lot of time, manpower and material resources to realize the classification management process by means of operation, and the classification of the same content is often relatively single, which cannot meet the needs of various classification aggregation, and is not conducive to users to quickly and accurately search for the content they want.

Images