Differential Evolution Dynamic Analysis in the Form of Complex Networks

In this work, we investigate the dynamics of Differential Evolution (DE) using complex networks. In this pursuit, we would like to clarify the term complex network and analyze its properties briefly. This chapter presents a novel method for analysis of the dynamics of evolutionary algorithms in the form of complex networks. We discuss the analogy between individuals in populations in an arbitrary evolutionary algorithm and vertices of a complex network as well as between edges in a complex network and communication between individuals in a population. We also discuss the dynamics of the analysis.

Characterization and Coarsening of Autonomous System Networks

This Chapter studies the correlations among well-known complex network metrics and presents techniques to coarse the topology of the Internet at the Autonomous System (AS) level. We present an experimental study on the linear relationships between a rich set of complex network metrics, to methodologically select a subset of non-redundant and potentially independent metrics that explain different aspects of the topology of the Internet. Then, the selected metrics are used to evaluate graph coarsening algorithms to reduce the topology of AS networks. The presented coarsening algorithms exploit the k-core decomposition of graphs to preserve relevant complex network properties.

On Vertex Cover Problems in Distributed Systems

An undirected graph can be represented by G(V,E) where V is the set of vertices and E is the set of edges connecting vertices. The problem of finding a vertex cover (VC) is to identify a set of vertices VC such that at least one endpoint of every edge in E is incident to a vertex V in VC. Vertex cover is a very important graph theoretical structure for various types of communication networks such as wireless sensor networks, since VC can be used for link monitoring, clustering, backbone formation and data aggregation management. In this chapter, we will define vertex cover and related problems with their applications on communication networks and we will survey some important distributed algorithms on this research area.

A Network Analysis Method for Tailoring Academic Programs

Producing or updating an academic program implies a significant effort: involving people, academic units, knowledge elements, regulations, institutions, industry, etc. Such effort entails a complexity related to the volume of elements involved, the diversity of the origins of contributions, the diversity of formats, the representation of information, and the required granularity. Moreover, such effort is a common task performed by humans who collaborate for long periods of time participating in frequent meetings in order to achieve agreement. New educational approaches are heading to adaptive, flexible, ubiquitous, asynchronous, collaborative, hyper-mediated, and personalized strategies based on modern Information and Communication Technologies (ICT). We propose an approach for tailoring academic programs to provide a practical and automated method to discover and organize milestones of knowledge through the use of Complex Networks Analysis (CNA) techniques. Based on indicators from CNA, the act of tailoring an academic program acquires meaning, structure and even body elements.

Application of Biomedical Image Processing in Blood Cell Counting using Hough Transform

In the field of biomedicine, blood cells are complex in nature. Nowadays, microscopic images are used in several laboratories for detecting cells or parasite by technician. The microscopic images of a blood stream contain RBCs, WBCs and Platelets. Blood cells are produced in the bone marrow and regularly released into circulation. Blood counts are monitored with a laboratory test called a Complete Blood Count (CBC). However, certain circumstances may cause to have fewer cells than is considered normal, a condition which is called “low blood counts”.This can be accomplished with the administration of blood cell growth factors. Common symptoms due to low red blood cells are:fatigue or tiredness, trouble breathing, rapid heart rate, difficulty staying warm, pale skin etc. Common symptoms due to low white blood cells are: infection, fever etc. It is important to monitor for low blood cell count because conditions could increase the risk of unpleasant and sometimes life-threatening side effects.

Triadic Substructures in Complex Networks

An important topological characteristic which has been studied on networks of diverse origin is the abundance of motifs – subgraph patterns which occur significantly more often than expected at random. We investigate whether motifs occur homogeneously or heterogeneously distributed over a graph. Analyzing real-world datasets, it is found that there are networks in which motifs are distributed highly heterogeneously, bound to the proximity of only very few nodes. Moreover, we study whole graphs with respect to the homogeneity and homophily of their node-specific triadic structure. The former describes the similarity of subgraphs in the neighborhoods of individual vertices. The latter quantifies whether connected vertices are structurally more similar than non-connected ones. These features are discovered to be characteristic for the networks' origins. Beyond, information on a graph's node-specific triadic structure can be used to detect groups of structurally similar vertices.

Design of Structural Controllability for Complex Network Architecture

Networks are all-pervasive in nature. The complete structural controllability of a network and its robustness against unwanted link failures and perturbations are issues of immense concern. In this chapter, we propose a heuristic to determine the minimum number of driver nodes for complete structural control, with a reduced complexity. We also introduce a novel approach to address the vulnerability of the real-world complex networks, and enhance the robustness of the network, prior to an attack or failure. The simulation results reveal that dense and homogenous networks are easier to control with lesser driver nodes, and are more robust, compared to sparse and inhomogeneous networks.

Evolutionary Computation Techniques for Community Detection in Social Network Analysis

Community detection is a topic of great interest in complex network analysis. The basic problem is to identify closely connected groups of nodes (i.e. the communities) from the networks of various objects represented in the form of a graph. Often, the problem is expressed as an optimization problem, where popular optimization techniques such as evolutionary computation techniques are utilized. The importance of these approaches is increasing for efficient community detection with the rapidly growing networks. The primary focus of this chapter is to study the applicability of such techniques for community detection. Our study includes the utilization of Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) with their numerous variants developed specifically for community detection. We have discussed several issues related to community detection, GA, PSO and the major hurdles faced during the implication of evolutionary approaches. In addition, the chapter also includes a detailed study of how these issues are being tackled with the various developments happening in the domain.

On Mutual Relations amongst Evolutionary Algorithm Dynamics and Its Hidden Complex Network Structures

In this chapter, we do synthesis of three partially different areas of research: complex networks, evolutionary computation and deterministic chaos. Ideas, results and methodologies reported and mentioned here are based on our previous results and experiments. We report here our latest results as well as propositions on further research that is in process in our group (http://navy.cs.vsb.cz/). In order to understand what is the main idea, lets first discuss an overview of the two main areas: complex networks and evolutionary algorithms.

Social Network Analysis

Social Network Analysis (SNA) looks at how our world is connected. The mapping and measuring of connections and interactions between people, groups, organizations and other connected entities are very significant and have been the subject of a fascinating interdisciplinary topic . Social networks like Twitter, Facebook, LinkedIn are very large in size with millions of vertices and billions of edges. To collect meaningful information from these densely connected graph and huge volume of data, it is important to find proper topology of the network as well as analyze different network parameters. The main objective of this work is to study network characteristics commonly used to explain social structures. In this chapter, we discuss all important aspect of social networking and analyze through a real time example. This analysis shows some distinguished parameters like number of clusters, group formation, node degree distribution, identifying influential leader/seed node etc. which can be used further for feature extraction.