Efficient Comparison of Multiple Complex Networks
Abstract
Network alignment (NA), one of the most popular network science/mining tasks, aims to compare networks corresponding to different systems in order to identify network regions of the systems (dis)similarities, thus allowing for learning something about a poorly understood system from a well understood system based on their aligned network regions. As such, NA has applications in a variety of domains, including computational biology, chemoinformatics, neuroscience, computational linguistics, artificial intelligence, computer vision, and web mining. Since complexity theory dictates that the problem of NA is computationally hard, this project introduced novel computationally efficient yet accurate heuristic NA approaches, such as those for alignment of multiple networks (as opposed to traditional pairwise NA), dynamic networks (as opposed to traditional static NA), or heterogeneous networks (as opposed to traditional homogeneous NA). The project resulted in 10 published or submitted papers and 16 conference presentations of the project results. It supported eight researchers (the principal investigator, a postdoctoral researcher, four Ph.D. students, and two undergraduate students).
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 06, 2019
- Accession Number
- AD1097136
Entities
People
- Tijana Milenković
Organizations
- University of Notre Dame