Media content association mining method based on event relationship discovery

Abstract

The invention provides a media content association mining method based on event relationship discovery. The method comprises the following steps: constructing a media content library; constructing and training a media content event extraction model, and identifying and extracting events contained in each piece of media content in the media content library; identifying and extracting event elements contained in each event; constructing a media content event association graph based on the extracted events and event elements; an event-based content relationship mining module is used to mine relationships between media contents, including a sequential relationship, a correlativity, a causal relationship and a supplemental relationship. According to the method, the events and the event elements in the media contents are extracted, multiple types of association relationships among the media contents are mined, the association contents are reasonably and effectively organized according to the relationship types among the media contents, and description and information expansion are performed on the target media contents from different angles; the associated content list with clear veins and high interpretability is displayed for the user, and the reading experience of the user is improved.

Description

technical field [0001] The invention belongs to the technical field of information retrieval, and in particular relates to a media content association mining method based on event relationship discovery. Background technique [0002] The existing media content association mining methods are generally divided into two categories: 1. Learning the semantic features corresponding to the media content through natural language processing, image recognition and other methods, and obtaining the associated content of the media content by calculating the semantic similarity of the media content. 2. Mining the relationship between media content based on media content tags. [0003] The problems of the above methods are as follows: On the one hand, using the semantic information similarity to mine the association relationship can only mine the associated content with similar semantics, ignoring the associated content that is not similar in semantics but logically related. On the other ...

AI Technical Summary

Problems solved by technology

[0003] The problems with the above methods are: On the one hand, using semantic information similarity to mine association relationships can only mine related content with similar semantics, ignoring related content with dissimilar semantics but logically related

On the other hand, using media content tags to establish the relationship between media content, the establishment of a media content tagging system requires a lot of human resources or there are problems with complicated tags and untargeted tags, and the related content that has been mined has limitations and cannot Digging into deeper relationships between media content

Finally, the relationship between the related content obtained by the above two types of mining methods is single, undirected, and indiscriminate, resulting in poor interpretability of the related relationship between the contents, and it is impossible to clearly understand the related content of the target content. Interpretive organization

Images