Description

In the era, where an enormous amount of data is getting generated from various

resources in different formats, it is highly required to categorize this data in proper format to

process useful knowledge which could be utilized effectively. Clustering technique is one of the

effective and popular techniques to segregate data by abstracting underlying structure of

the data. This approach is used to organize the data either to form a group of

individuals or categorize as a hierarchy of groups. Clustering becomes an important technique

to analyze large amounts of data which is frequently applied in various domains of

engineering, science and other well-known areas such as biology, marketing, psychology,

medicine, remote sensing, computer vision etc. The representation of data that has been

done in clustering analysis is then undergone the observation. It is done to articulate and

justify the grouping of data. The investigation is carried out to see whether the phenomenon

of clustering is fitting into the preconceived ideas and experiments.

Data mining is the domain where data is being retrieved, processed and transformed

into information. In data mining, clustering is one of the most frequently used forms of

exploratory data analysis which belongs to unsupervised classification of patterns

into groups .Clustering works as to divide data into groups on the basis of similarity

and dissimilarity . It is the collection of those data sets and entities which lies in these

groups pertaining to similar and different properties.

In most of the cases, clusters are formed by exploring their internal homogeneous properties and

external separation of dataset. In prescribed clusters, patterns are found to be similar in the

same groups and different in different groups . Data analysis belongs to many

computing applications; it is considered to be involved in the design phase or as a

part of their online functions. Data analysis procedures can be categorized as either

exploratory or confirmatory, based on the models which are appropriate for the source of the

data, but a key element in both types of processes is considered to be grouping, or classification of measurements.