Prof. K. Selçuk Candan
Smart Data Services for Sensemaking in Human-Centered Dynamic Systems.
Abstract: Many socio-economical critical human-centered domains (such as sustainability, public health) are characterized by highly complex and dynamic systems, requiring data and model driven situational awareness and decision making. Successfully tackling many urgent challenges in these domains requires obtaining a deeper understanding of complex relationships and interactions among a diverse spectrum of entities in different evolving contexts. Models have to be constructed in the presence of sensed data, along with applicable physical models, from multiple sources, often characterized by varying levels of coverage and accuracy. Moreover, both data and models required for the said situational awareness and predictions are defined over high-dimensional and time-varying parameter spaces and require causally informed analysis within the appropriate context. Thus, operations in these domains necessitate addressing several major challenges, including latent contexts of impact, heterogeneous networks of entities, dynamicity of impact in varying contexts, and high computational and I/O costs of context-sensitive impact discovery. These algorithms and the novel data platforms they are deployed in need to be efficient and scalable in terms of off-line and on-line running times and their space requirements. In this talk, we will provide several examples from National Science Foundation and Department of Energy funded projects on resilient building energy systems and discuss outlines of possible computational approaches to these challenges.