SBIR/STTR Award attributes
US military and intelligence agencies, including the NGA, have invested significant resources in data collection and effective search and analytics tools. However, due to increasing amounts of data, finding relevant information has become more difficult. NGA analysts are often overwhelmed with data and spend too much time solving data problems rather than solving intelligence problems. Thus, there is an important need for recommender system technology that pushes relevant un-queried data to analysts through automation and machine learning techniques. In Phase I, Numerica developed an innovative, domain-agnostic, knowledge-graph-based collaborative recommender system (KCRS) for spatio-temporal intelligence documents. Key features of the KCRS prototype include (i) a multi-layer knowledge graph-based recommender to accommodate different data models and algorithms, while supporting real-time updates and computations at scale, (ii) deep learning for probabilistic linking of documents based on content to enable relevant document recommendations, (iii) identification of users performing similar tasks to enhance collaboration, and (iv) potential for exploiting user feedback for persistent improvements of recommendations over time. In Phase II, Numerica will further develop, mature, and demonstrate the KCRS prototype to NGA analysts and key stakeholders. In addition, Numerica will leverage its real-time commercial space situational awareness (SSA) platform and its in-house team of SSA analysts in order to mature, deploy, and demonstrate the KCRS technology on real-world use cases.
