This chapter is from the book
This chapter is from the book
This chapter is from the book
1.4 Challenges in Studying Complex Networks
While complex network approach can be very useful for various areas of research and application, many challenges exist in its use. Following are the major challenges of studying complex networks:
High data volume Most of the real-world complex networks involve large numbers of nodes. Further, the complex presence of edges in the network results in huge data volume. For example, the Internet has many billions of nodes and edges. Storing and manipulating such large amounts of data requires sophisticated storage resources. Further, based on the approach used for analyzing the complex network dataset, intermediate processing required for data storage can be higher. For example, approaches such as graph signal processing require computation of eigenvectors and eigenvalues, which demand significant storage space.
Complexity in mapping physical systems to realistic complex network model Another major issue in complex network analysis is the difficulty in translating a real-world problem to a complex network model. In many real-world cases, only limited information about the nodes and edges of the physical world system can be captured to the complex network model. For example, consider a protein-protein interaction network where the edges are modeled on the basis of very limited interactions between proteins. In reality, the protein-protein interaction is far more complex, and it is challenging to model the complexity to the actual detailed level.
High computational complexity In order to characterize a real-world complex network, one needs a significant amount of computing resources. Therefore, algorithms must be efficient as well as able to process large-sized networks. Running complex network algorithms over large complex network datasets is computationally expensive. Approaches such as graph signal processing can take significant amounts of computing power to carry out complex network analysis of large real-world graphs.
