TY  - JOUR
T1  - Real World Data Clustering using a Hybrid of Normalized Particle
Swarm Optimization and Density–Sensitive Distance Measure
AU - Fagbola, Temitayo AU - Oludayo, Olugbara AU - Thakur, Surendra 
JO  - Journal of Engineering and Applied Sciences
VL  - 14
IS  - 17
SP  - 6317
EP  - 6335
PY  - 2019
DA  - 2001/08/19
SN  - 1816-949x
DO  - jeasci.2019.6317.6335
UR  - https://makhillpublications.co/view-article.php?doi=jeasci.2019.6317.6335
KW  - k–means
KW  -normalized particle swarm optimization
KW  -clustering
KW  -real world dataset
KW  -density–sensitivedistance metric
KW  -min–max normalization
AB  - k–means is among the most widely used classical partitioned clustering algorithms mainly because
of its quick convergence rate, adaptability nature to sparse data and simplicity of implementation. However, it
only guarantees convergence of sum of square’s objective function to a local minimum while its convergence
to global optimum appears NP–hard when introduced to large, noisy and non–convex structures. This in turn
maximizes its error margin. Most currently existing improvements on k–means adopt techniques which further
introduce additional challenges including inaccurate clustering results, high space and time complexities and
sometimes premature convergence on k–means. However, high accuracy with large datasets, robustness to
noisy data, low clustering time and low sum–of–squared error are sought–after capabilities of good clustering
algorithms. In this study, a hybrid Normalized Particle Swarm Optimized–Density Sensitive (NPSO–DS) k–means
algorithm is developed to manage the aforementioned limitations of k–means. The proposed NPSO–DS k–means
algorithm combines the global stability feature of the normalized Particle Swarm Optimization (PSO) technique
incorporating a min–max technique and a clustering error as objective function with the stable properties of a
density–sensitive k–means to realize convergence of particles to global optimum with large and noisy real–world
datasets. Using clustering accuracy, sum–of–squared error and clustering time as performance metrics, the
experimental evaluation results obtained when the developed algorithm was tested on Educational Process
Mining (EPM) and wine datasets indicate that it is significantly capable of consistently yielding high quality
results. Furthermore, the developed NPSO–DS k–means algorithm could identify non–convex clustering
structures and offers appreciable robustness to noisy data, thus, generalizing the application areas of the
baseline k–means algorithm.
ER  -