Data Fusion Support to Activity-Based Intelligence

This new resource provides a coherent, intuitive, and theoretical foundation for the fusion and exploitation of traditional sensor data as well as text-based information. In addition to presenting a detailed discussion of base-level data fusion requirements, a variety of higher level exploitation algorithms are presented that perform fully automated relationship discovery, rank interest level of entities, and support context-sensitive behavior understanding (both static and dynamic context). This book identifies eight canonical fusion forms as well as twenty foundational fusion services to enable formal mapping between models and services. Normalization and representation processes for (hard) sensor data and (soft) semantic data are described as well as methods for combining hard and soft data. Included is a prototype fusion system developed to implement virtually all the presented applications in order to demonstrate the robustness and utility of the design principles presented in this resource. The prototype system presented supports a variety of user workflows and all the applications are fully integrated.

Introduction and Background; Foundational Theory; Prototype Implementation; Higher Level Reasoning; Special Functions; The Way Ahead; Refinement Extensions, Binary Fusion, Putting it all Together: Implementing the Analytic Workflow; Reasoning Paradigms; Spatial Reasoning Support; Temporal Reasoning Support; Summary and Next Steps.