Copyright multi-granularity splitting and commercial value mining method

Abstract

The invention belongs to the technical field of artificial intelligence, and further relates to a copyright multi-granularity splitting and commercial value mining method. The method comprises a copyright multi-granularity splitting method and a copyright commercial value mining method. The copyright multi-granularity splitting method comprises scene angle splitting and specific person splitting. The multi-granularity splitting method for the copyright further comprises full-text-oriented language feature mining and scene feature mining, character feature mining and character relationship feature mining which are based on scene angle splitting. The mining method for the copyright commercial value comprises copyright topic mining, copyright quality judgment and copyright commercial value pricing. The method can solve the problems that in the prior art, machine identification cannot correctly judge the quality of copyrights, high-quality copyrights are prone to being missed, a copyright quality evaluation report based on automatic formation cannot be achieved, and a large-scale copyright incubation platform cannot be constructed.

Description

technical field [0001] This application belongs to the technical field of artificial intelligence, and further relates to the multi-granularity splitting of copyright and its commercial value mining method. Background technique [0002] At present, the main ways to discover high-quality copyrights are: new copyrights of famous authors; manual review by professionals, such as book planners of publishing companies and project developers of film and television companies, who devote a considerable part of their energy to discovering high-quality copyrights; Trial reading, such as various web sites, will allocate part of the traffic to new web texts, and then determine the traffic allocation quota for the copyright based on subsequent reading depth and word-of-mouth. However, on the one hand, the existing machine recognition effect is not good, and there is still a gap compared with manual recognition. Using machine recognition is easy to miss high-quality copyrights; the output ...

AI Technical Summary

Problems solved by technology

[0003] The main problem solved by this application is to provide multi-granularity splitting of copyright and its commercial value mining method to solve the incomplete splitting of copyright in the prior art when machines identify copyright, resulting in incomplete understanding of copyright, thus As a result, it is impossible to make correct quality judgments on copyrights, and it is easy to miss high-quality copyrights. At the same time, there is no commercial value evaluation of copyrights in the existing technology, so it is impossible to realize copyright quality evaluation reports based on automatic formation, and it is impossible to build large-scale Copyright incubation platform. At the same time, the cost of manual identification and review of non-machine identification is extremely high, and compared with tens of millions of new copyrights, the number that professionals can handle is extremely limited. At the same time, because there is no copyright review Correct quality judgment, if tens of millions of new copyrights are allocated to users for trial reading, it will cause great user harm and other problems