Resolving Independent Perspectives by Providing Learning-Enabled Enhanced Fusion For Elastic Cloud Technologies (RIPPLE-EFFECT) Navy SBIR FY2013.2 Sol No.: Navy SBIR FY2013.2 Topic No.: N132-135 Topic Title: Resolving Independent Perspectives by Providing Learning-Enabled Enhanced Fusion For Elastic Cloud Technologies (RIPPLE-EFFECT) Proposal No.: N132-135-0650 Firm: DECISIVE ANALYTICS Corporation 1235 South Clark Street Suite 400 Arlington, Virginia 22202 Contact: Mark Frymire Phone: (703) 414-5139 Web Site: http://www.dac.us Abstract: To maintain situational awareness, analysts must sift through and fuse information across multiple documents, data sources, and modalities (text, imagery, and biometrics). The emergence of Big Data has placed an enormous burden on the analyst as the volume of data to examine has increased dramatically while the analyst's capacity to understand and fuse information remains constant. Additionally, the data remains distributed across geographically separated systems with low-bandwidth connections. The analyst is presented with an incomplete data space from which to produce an intelligence picture. DAC proposes a system called Resolving Independent Perspectives by Providing Learning Enabled Enhance Fusion For Elastic Cloud Technologies (RIPPLE-EFFECT). RIPPLE-EFFECT provides a framework to fuse semantic enhancements from multiple vendors through the use of machine learning algorithms which infer correlations between both the semantic structure and the extracted semantic content. RIPPLE-EFFECT supports scalable cross-document inference over the semantically enhanced data space with dynamically evolving search patterns based on data encountered during the search and the semantic meaning behind the initial query. RIPPLE-EFFECT maintains a consistent data space and intelligence picture across geographically separated systems specialized to the area of interest and time period of interest for each system through lock-free continuous synchronization. Benefits: The proposed system called Resolving Independent Perspectives by Providing Learning Enabled Enhance Fusion For Elastic Cloud Technologies supports the automation of introducing new sources of semantic enhancement through ontology-agnostic data fusion. The data fusion is implemented as Hadoop MapReduce jobs to support the massive volume of data. The cross-document inference algorithm is implemented utilizing the Accumulo database to support low-overhead scalable search. The combined capability reduces the burden on the analyst to fuse data across the multitude of documents, sources, and modalities. Return