Efficient Representation, Measurement, and Recovery of Large-scale Networks

Real-world networks have millions of users and are complex in structure. Efficient techniques are required to capture the network characteristics as compact data. The overall purpose of this research is to mine information from partially or completely available graph data and obtain optimum data representation for networks. The first phase involves studying network data to draw meaningful relations and properties about the network. We extend this work to introduce a novel way of sampling graphs in lossless manner. The second phase involves observing and processing the partial data available to complete the graph data by estimation methods. We also leverage our knowledge about graphs to design an optimal network sampling technique. Subsequently, these techniques will be applied to both static as well as dynamic graphs.