Content Repurposing Platform Utilizing Metadata Extracted from Rich Media

The amounts of unstructured data stored in companies are rapidly growing up, and the needs for utilizing those data are coming up. We've been working on research and development of a platform that can ease utilization of unstructured data. In order to make situation aware information system in organizations such as hospitals, utilizing unstructured data such as texts, images, and voices, in combination with structured data is required. For example, most of the hospitals own several information systems for each purpose and each organization. Thus, searching patient information across multiple data sources or data types is difficult problem that must be solved. To solve this problem, we developed content repurposing platform that enables search refinement across data sources using metadata database which store metadata extracted from each data sources. The platform is constructed on top of object storage and includes metadata extraction engine, metadata DB, content router, and search engine. We applied the content repurposing platform to a hospital and made preliminary experimental evaluation. The result of experiment shown that the time to prepare data for a case conference was dramatically reduced to 1/68 meaning that most of the manual operations for data preparation could be removed.