SC Colloquium: "From Data Mining to Knowledge Mining in Smart Infrastructure"
Given the tremendous production of data in smart grids and smart cities, there is a need for new powerful tools that can automatically generate useful knowledge from a variety of data, and present it to systems operators and stakeholders. In an effort to create knowledge mining tools, researchers have been exploring methods and algorithms developed in machine learning, statistical inference, information theory, pattern recognition, network modeling, and operations research communities. The first part of this talk will address a compendium of ideas on the applicability of sensor networks, and machine learning tools for data mining and data fusion. The effective use of causality inference and information theory makes it possible to develop knowledge mining tools that derive useful new knowledge not only from large amounts of data but also from limited and weakly relevant data sets. The second part of the talk outlines our current research direction on smart grids and smart cities by which we characterize the interdependency and interconnectivity of electricity networks with other urban utility networks around the novel concept of co-mobility.