SBIR/STTR Award attributes
AvaWatz’s SOPHIA is an artificial intelligence (AI) based Analytics and interactive Visualization software that reduces the cognitive burden on intelligence analysts tasked with interpreting large volumes of complex Intelligence-Surveillance-Reconnaissance (ISR) data. The ability to collect ISR data has far outpaced the ability to understand it. Technology has greatly amplified our capacity to collect raw facts, but it has not made equal progress in assisting our understanding of those facts. While we can accumulate vast quantities of observations, it is still the responsibility of human analysts to attach meaning to those observations, to find patterns and connections that bind events together into narratives, and to select the rare scenarios with tactical or strategic significance. Cognitive overload, due to natural limits on human attention and perception, is the principal bottleneck that SOPHIA solves to fully exploit intelligence collection. SOPHIA’s Analytics and Visualization components treat connected data elements, such as sequences and narratives, as the basic units of data to be analyzed and displayed, and semantic features are used to assist the analyst’s sensemaking process. SOPHIA uses a unique blend of machine learning, reasoning and reinforcement models that sharpen its results and reduces its rate of false positives.
