Measurement and comparison of clustering algorithms

Within this report, a variety of clustering algorithms are described in addition to their workings explained. These are in comparison to one another by implementing them on quantity of graphs having a known architecture.These clustering algorithm, inside the order they’re implemented, are listed below: Nearest neighbour hillclimbing, Nearest neighbour big step hillclimbing, Best neighbour hillclimbing, Best neighbour big step hillclimbing, Gem 3D, K-means simple, K-means Gem 3D, One cluster then one cluster per node.The graphs are Unconnected, Directed KX, Directed Cycle KX and Directed Cycle.The outcomes of those clusterings are compared to each other well based on three criteria: Time, Quality and Extremity of nodes distribution. This permits us to discover which algorithm is the most appropriate by which graph. These artificial graphs are then in contrast to the reference architecture graph to achieve the conclusions.

Contents: Measurement and comparison of clustering algorithms

1. INTRODUCTION
2. CLUSTERING ANALYSIS
3. CLUSTERING ALGORITHMS
3.1 HILLCLIMBING ALGORITHM
3.1.1 Nearest neighbour
3.1.2 Nearest neighbour big step
3.1.3 Best neighbour
3.1.4 Best neighbour big step
3.2 GEM 3D
3.3 KMEANS
3.3.1 K-means simple
3.3.2 K-means Gem3D
3.4 ONE CLUSTER
3.5 ONE CLUSTER PER NODE
4. ENVIRONMENT OF IMPLEMENTATION ALGORITHMS AND TESTING
4.1 WHAT IS ECLIPSE AND ECLIPSE FOUNDATION?
4.2 HISTORY OF ECLIPSE
4.3 DOWNLOADING ECLIPSE
4.4 PROJECT CODES
5. TABLES AND DIAGRAMS
5.1 TIME TABLES AND DIAGRAMS
5.1.1 Nearest neighbour
5.1.2 Nearest neighbour big step
5.1.3 Best neighbour
5.1.4 Best neighbour big step
5.1.5 Gem 3D
5.1.6 K-means simple
5.1.7 K-means Gem3D
5.1.8 One cluster
5.1.9 One cluster per node
5.2 QUALITY TABLE AND DIAGRAMS
5.2.1 Nearest neighbour
5.2.2 Nearest neighbour big step
5.2.3 Beat neighbour
5.2.4 Best neighbour big step
5.2.5 Gem3D
5.2.6 K-means simple
5.2.7 K-means Gem3D
5.2.8 One cluster
5.2.9 One cluster per node
5.3 EXTREMITY
5.3.1 Nearest neighbour
5.3.2 Nearest neighbour big step
5.3.3 Best neighbour
5.3.4 Best neighbour big step
5.3.5 Gem 3D
5.3.6 K-means simple
5.3.7 K-means Gem 3D
6. REFERENCE GRAPH
6.1 REFERENCE GRAPH TIME, PRECISION, RECALL AND QUALITY
6.2 REFERENCE GRAPH EXTREMITY
7. COMPARING REFERENCE GRAPH WITH ARTIFICIAL GRAPHS
7.1 TIME
7.1.1 Nearest neighbour
7.1.2 Nearest neighbour big step
7.1.3 Best neighbour
7.1.4 Best neighbour big step
7.1.5 Gem 3D
7.1.6 K-means simple
7.1.7 K-means Gem 3D
7.1.8 One cluster….