How shall we build the network from big data,
without losing important information about higher-order dependencies?
Network-based representation has quickly emerged as the norm in representing rich interactions among the components of a complex system for analysis and modeling: movement of hundreds of thousands of ships form a global shipping network, powering the transportation and economy while inadvertently translocating invasive species; interactions of billions of people on social networks, facilitating the diffusion of information.
Given the ship trajectories, to construct a global shipping network, the conventional approach is to count the number of voyages between port pairs as edge weights in the network.
What are the advantages of using HON
instead of the conventional first-order network or the fixed second-order network?
Jian Xu, Thanuka L. Wickramarathne, and Nitesh V. Chawla. "Representing higher-order dependencies in networks." Science Advances (2016)
Code for generating Higher-order Network (HON) from data with higher-order dependencies.
HONS @ NetSci 2018
Jun. 12, 2018. Paris, France
Prof. Nitesh Chawla will contribute a talk on the latest developments of HON, including the improved parameter-free algorithm, its application in anomaly detection in sequential data, and modeling species invasion in the Arctic.
Dr. Jian Xu (Lucy Family Institute for Data & Society alumni) will host the morning session “Beyond Markov models”.