Data Mining and Knowledge Engineering

N. Murali ¹, K. Kumaran ¹

Affiliations
1 Computer Science and Engineering, E.G.S. Pillay Engineering College, Nagapattinam, IN

Abstract

The explosive growth in the number of mobile devices such as Internet-enabled cellular phones, wireless handheld devices, wireless laptops, and tablet PCs has driven the corresponding growth in applications for mobile computing. These applications usually belong to one of two classes: collaborative applications and individual application. While collaborative applications require several mobile devices to work together and include peer-to-peer computing and grid computing, individual applications are local to the mobile device and its user. In this paper, we present a framework that allows mobile devices to collaboratively work on a computationally-expensive problem. Such a problem is decomposed into smaller tasks and distributed across other mobile devices willing to share their computational power with others. This framework is based on the paradigm of grid computing applied to the domain of wireless mobile devices. This paper presents our current implementation of the framework architecture and simulation of its functionality. Also, we focus on the issues of mobility, QoS and network stability and their effect on the performance of our collaborative problem-solving framework.

Keywords

Grid Architecture, Wireless Network, Mobile Devices.

Full Text

     

M. A. Arunagiri ¹, K. Kumaran ², N. Murali ²

Affiliations
1 C.C.M.R. Polytechnic College, Thanjavur-613003, IN
2 E.G.S. Pillay Engineering College, Nagapattinam, IN

Abstract

A "graph" refers to a collection of vertices or nodes and a collection of edges that connect pairs of vertices. A graph may be undirected, meaning that there is no distinction between the two vertices associated with each edge, or its edges may be directed from one vertex to another. Spanning Tree is a tree that connects all vertices in the graph. In a single graph, we can have many number of spanning tree. A minimum spanning tree (MST) is then a spanning tree with edge cost less than or equal to the cost of every other spanning tree. Several algorithms have been proposed to find minimum spanning tree. Considering the entire graph to construct MST is a difficult task since the number of nodes increases then the complexity also increases and is vice versa. So here we apply divide and conquer approach to split the nodes into several groups (clusters) and then MST is performed to each cluster and finally clusters are connected. Several clustering algorithms are used for distance based clustering here we prefer to take hierarchical clustering algorithm to find minimum spanning tree. Present work the authors conclude that performance of hierarchical clustering is better than that of other techniques. The experimental results of the performance outcome of the hierarchical are so better that the K mean clustering technique.

Keywords

Hierarchical Clustering, MST, K Means Clustering.

Full Text