IASA 2018 Annual Conference

Based on a digital media archive management platform, this paper studies the ontology model of media data, reviews the adaptive technical methods and tools according to the required entity relations, establishes a media Knowledge Graph, and designs a prototype system based on the Knowledge Graph Retrieve to automate the intelligent retrieval of associated media assets. Based on the annotation data provided by the existing digital media files, this article manages the data, creates the media ontology, and defines the corresponding ontology concepts and relation sets. It then details efforts to analyze the data format of the existing digital media archives management platform, to realize the text data extraction technology, and to extract the entity and entity relations from it to establish the Knowledge Graph. Compared with traditional keyword-based full-text search, the retrieval method based on the Knowledge Graph can effectively improve the quality of search and resource arrangement, application, and efficiency. Especially in the fields of digital archives and cultural media, resource management and search results can be knowledge-based and can provide great practical value.