Intra graph clustering using collaborative similarity measure

Abstract

Graph is an extremely versatile data structure in terms of its expressiveness and flexibility to model a range of real life phenomenon. Various networks like social networks, sensor networks and computer networks are represented and stored in the form of graphs. The analysis of these kind of graphs has an immense importance from quite a long time. It is performed from various aspects to get maximum out of such multifaceted information repository. When the analysis is targeted towards finding groups of vertices based on their similarity in a graph, clustering is the most conspicuous option. Previous graph clustering approaches either focus on the topological structure or attributes likeness, however, few recent methods constitutes both aspects simultaneously. Due to enormous computation requirements for similarity estimation, these methods are often suffered from scalability issues. In order to overcome this limitation, we introduce collaborative similarity measure (CSM) for intra-graph clustering. CSM is based on shortest path strategy, instead of all paths, to define structural and semantic relevance among vertices. First, we calculate the pair-wise similarity among vertices using CSM. Second, vertices are grouped together based on calculated similarity under k-Medoid framework. Empirical analysis, based on density, and entropy, proves the efficacy of CSM over existing measures. Moreover, CSM becomes 123 Distrib Parallel Databases a potential candidate for medium scaled graph analysis due to an order of magnitude less computations.

A Graph Clustering Algorithm Using Attraction-Force Similarity for Community Detection

“…IGC-CSM [25] uses Jaccard coefficient to define the structural similarity between nodes. It divides the connection relationships among all nodes in a weighted categorical graph into three types, which are direct connection, indirect connection and non-connection.…”

“…And when there are many attributes, the time complexity of the algorithm will be high. Nawaz et al [25] proposed IGC-CSM algorithm. It considers both of network topology and node characteristics comprehensively.…”

A graph clustering method for community detection in complex networks

“…It combines the paradigms of dense subgraph mining and subspace clustering to obtain sets of objects that are densely connected within the associated graph and also show high similarity regarding their attributes. Nawaz et al [37] proposed IGC-CSM, which also combines structural and attribute aspects, and utilized the K-Medoids [38] framework for clustering. This approach is simple for similarity measures, but it is difficult to scale up for large graphs.…”

LED: A fast overlapping communities detection algorithm based on structural clustering

“…Figure 3(a) shows the Karate network, with the instructor and the administrator represented by vertex1 and vertex34, respectively. The weights of edges are generated by calculating their structural similarities using formula (2). We set the threshold α = 0.5.…”

“…Community structure [1] is a common property of many networks. In terms of our experience on online social networks, it's a common sense that people with same interests trending to get together as communities: subsets of vertices within which vertex to vertex connections are dense, but between which connections are sparse [2,3,5,6]. Also, some active members of social network may take part in many communities simultaneously, which in networks we call them overlapping vertices.…”