Designing Large Data Handling Architectures

Global War on Terror requires critical need for accessing intelligence information databases for actionable decision processes. To enhance intelligence information based decision making a methodology must be established for storage and retrieval from multiple database systems, and subsequent analysis based on information’s meaning rather than predetermined manually assigned categories. Open and standards based architectures are needed to efficiently assemble large amounts of data with greater agile information sharing strategies. Automation of handling large data amounts can be achieved by using metadata, alignment of vocabularies, data sharing governance rules, and defined business processes. Analatom proposes investigating a more robust query and index paradigm having large data handling architectures focusing on ‘Concept Footprints’. Associated pools of data leave ‘multiple’ tracks of variously weighted associations through historical usage and prior user interest. Proposed software will allow multidimensional associations to form (organize and attract to similar concepts and associations) within like concept neighborhoods. These resulting (task oriented) multiple architectures are referred to as ‘Information/Knowledge Cubes’. These then afford access into extremely large data sets and repositories to be concept oriented. Additionally, queries need not be specific or limiting, but rather can be presented to data repository (Knowledge Cube) as incomplete or ‘fuzzy’ textural queries.