chapter{Conclusions}% and Future Research Directions} label{final:conclustion} drop{S}{ocial} networks have been the pulse of human society, traffic on these social networking sites currently rival that of the traditional Web. The digital human activity on social networking sites has generated a tremendous amount of data and due to the availability of affordable and portable digital devices, the social interactions or activities on these networking sites can be measured and stored in digital formats. These data contain a lot of important information and had been used in testing numerous sociological hypothesis and also modeling systems that depend on human activities and research in this area is actively ongoing in the research community. The social interaction as the result of the human activities on these networks generate a non-trivial topology in the process of time. %After the discovering of the ``heavy tail distribution", the small world phenomenon and high clustering in real networks, And various graph-theoretical tools have been developed and used to analyze this network topology and also modeled networks with such topology. In this dissertation, the theory and key technology challenges within social networks are thoroughly investigated and various novel findings are presented. The remaining parts of this chapter are organized as follows. The work and results obtained in the thesis are presented in Section ef{summary}. Section ef{ThesisContr} highlights the main
Social Networking is basically a map of relationships between individuals. It indicates ways in which they are connected through various social familiarities -- friends, close friends, school mates, casual acquaintance, family, business, etc. The theory views social relationships in terms of nodes and ties . Nodes are individuals in the network while ties are the relationships between them.
As we see in the different texts, social network has a problem. The problem is how we are going to use the social network in a right and normal way.
This thesis is separated into two main sections, namely a theoretical part and a practical part.
Following techniques used for data analysis, pre-processing, and data interpretation processes in the data analysis subject. The survey illustrates different data mining techniques used in mining diverse aspects of the social network over centuries going from the historical techniques to the up-to-date models, including our novel technique. Searching social networks can help us better understand how, we can reach other people. Adding to this research on small worlds, with their relatively small part between nodes, can help us design networks that facilitate transmission of data or other resources without having to exhale the network with many duplicate connections. Recent studies have shown that the nodes’ degrees, that is, the number of edges inclined to each node, and the distances between a pair of nodes, as captured by the shortest path length.
Social-graph based techniques represent the graph as \textbf{G=(V,E)} where G represents the graph, V represents the nodes in the graph and E represents the edges of the graph.\cite{gen5graph} The social- graph based mechanisms first divides the node set V into two disjoint sets- one which are real users and the second set which consists of likely Sybil users. The Sybil set consists of all the Sybil nodes and the edges between them and the non-Sybil region consists of the non-Sybil nodes and all the connections between the real nodes. It is assumed that the non-Sybil region of the network is tightly connected and has dense connections among its nodes as they trust each other. Hence the time taken to form connections between nodes, also called the mixing time of the region, is very small when compared with time taken to mix with the Sybil region.\cite{gen5graph} For example, the probability that a real user accepts a friend request from another real user in Facebook is much higher than the probability of the request being accepted from an unknown or fake user. Thus Sybil accounts find it difficult to build connections with the real user. This results in limited number of edges between the nodes in non-Sybil and the Sybil region. These edges are also called attack edges.
It’s a new age of technology; use of social networking websites is increasing day by day. Every single aspect of our lives asks for technological services. Technology is getting more important day by day for humans to survive, every single person now a day’s uses social networking websites like facebook, twitter, pinterest, instagram or at least one of these websites mainly facebook. Social networking is a way to connect with people with similar tastes as oneself and to communicate with friends and family that are far away.
According to Nicholas Christakis, social networks are like the pairs of people connected. Through his studies, he realized that socially connected people get embedded in other sorts of relationships such as marriage, friendship, or even spousal. He discovered that these connections are vast, and all of us are embedded in the broader set of relationships with each other. Therefore, social networks are the complex things of beauty that are so elaborate, so sophisticated and so ubiquitous.
A social network is a set of people (or organizations or other social entities) connected by a set of social relationships such as friendship, co-working or information exchange. Social networks are connected through various social familiarities ranging from casual acquaintance to close familiar bonds. Social network analysis provides both a visual and a mathematical analysis of relationships. Social network analysis (SNA) is a quantitative analysis of relationships between individuals or organizations. By quantifying such social structures it is possible to identify most important actors, group formations or equivalent roles of actors within a social network. This paper presents various properties or analysis measures for social
exttt{adjacency matrix}. The eigenvalue spectrum of complex networks provides information about their structural properties. There has been much research work on spectral properties of uncorrelated random graphs, however, there is less information on the spectral properties of social networks or sparse complex networks. We, therefore, want to address this research gap by analyzing the spectral properties of several real-life networks. The singular value distributions of selected social networks are shown in Figure~
In order to understand the structure of a complex social network, three random graph models have been used [10]. For an overall comparative study, the basic centrality measures, clustering coefficient, average path length, and the degree distribution were studied for the original network model as well as for the random graph models.
The purpose of this chapter is to give a brief introduction about this work and to introduce various
There are several key concepts at the heart of network analysis that are fundamental to the discussion of social networks that should be clarified for discussing social networks and social network data that are required in the network analysis. These concepts are actor, relational tie, group, relation, and network. The first concepts, actors are discrete individual, corporate, or collective social units that applicably focus on collections of actors that are all of the same type, such as member of entrepreneurs’ group, or an organisation (Wasserman & Faust, 1994). However, some methods allow looking at actors of conceptually different types or levels, or from different
In this section we present our computational results of the properties of the proposed socio-spatial network models and compare them with those of ER, SW, and BA networks when it is appropriate. More specifically, we measure clustering coefficient, network density, average path length, assortativity, transitivity, and min, max, mean, and standard deviation degree. While several algorithms have been proposed for some of these measures, for simplicity of reproducing the results, we use the algorithms that have been implemented in the Python’s Networkx library. All results are averaged over 25 independent model runs with different random seeds. It should be noted that in all analysis in this study we only consider undirected networks. Moreover, since the focus of this study is on the human social network, we limit our analysis to Onnela et al’s (2011) findings that α ≈ 1.5 and only consider three values of α = 1.2, 1.5, and 2 for our sensitivity analysis purposes.
Another form of attack on online social networks is the Social Link Disclosure based attacks, in which the goal of an attacker is to obtain knowledge of a significant fraction of the links in the network. The value of participating in an online social network for a user lies majorly in the ability to leverage the structure of the social network graph. But, knowledge of this social graph by parties other than the service provider paves the way for powerful data mining, some of which may not be desirable to the users. The ability to conduct these analyses on social networks is becoming increasingly important in the data mining, database and theoretical computer science communities [4.32]. Since nodes in social
The rest of this paper is organized as follows : in Section 2 we first briefly review the related works and after that propose our framework . In Section 3, we present our plan and methods for this research and finally, Section 5 talks about the outcome of the