MetaCORE: Metadata automated Categorization and Optimized Relevance Exploration

To support net-centric warfare, the Air Force must automate the generation and maintenance of metadata about both new and legacy information products. Metadata will enable Warfighters to retrieve necessary information quickly enough for accelerated ops tempos from among the rapidly increasing number of accessible information products. The metadata must include standard, domain-defined, user-generated, and automatically-generated attributes. We will build the MetaCORE (Metadata automated Categorization and Optimized Relevance Exploration) application as a complete metadata system architecture that can automatically populate a central repository with metadata mined from new and non-standardized legacy information products. A key advantage of MetaCORE is that all metadata will be stored in RDF, a next generation web standard accompanied by a powerful query language that provides greater functionality over other metadata storage formats. The modular MetaCORE system will enable the creation, filtering, retrieval and discovery of metadata using best-of-breed technology. Foremost among the technology that we will initially use to ensure that MetaCORE can achieve the most difficult task of discovering meta-data types and populating textually related metadata such as categories and keywords are Probabilistic Latent Semantic Analysis (PLSA) and Probabilistic Support Vector Machines (PSVM).