Explain the Cluster analysis, how it is different factor analysis and multi-dimensional scaling? Name a real-life company has used the Cluster analysis technique to address a business problem and how might that same technique be used at your own organization?
// Multivariate analysis is described as the analysis of several random correlated variables or measuring the quantitative variables. One aspect of this kind of analysis refers to dimensionality techniques that include multidimensional scaling, factor analysis, or cluster analysis. This paper will include the description of cluster analysis, and how it is different from the other two multivariate techniques that are factor analysis and multidimensional scaling. It will also contain an example of real life company, using this technique to solve their problems//.
Multivariate analysis is a technique to examine the relationship between independent variables and metric dependent variables. It examines a dependence or relationship between a set of dependent measures across a set of groups. Basically, it is used in experimental designs with the help of its different techniques. Cluster analysis, factor analysis and multidimensional scaling are also used for this purpose. Cluster analysis is a set of statistical methods that are used to classify the same groups or samples having similar characteristics. It is also referred to define look-a-like groups (Richarme, 2001).
To classify the look-a-like groups, a simple mechanism is followed that refers to the measurement of samples on the basis of their similarity and differences. The similar and different characteristics of the groups provide interchangeable words. Cluster analysis is totally based on the segmented method that refers to the discovery against prediction or predefined learning sets. All the samples that belong to the cluster are considered equal (Slezak, 2011). Clustering is the technique for statistical data analysis that can be used in several fields including pattern recognition, information retrieval, machine learning, image analysis, and bioinformatics (Richarme, 2001).
The characteristics of a cluster vary between one of many decisions to be taken, while choosing appropriate algorithm for a particular problem. Clusters have different algorithms according to the data objects and their properties. There are various cluster models to understand the key differences between various algorithms. It includes connectivity models that are based on hierarchy, graph based models, group models, density models, distribution models and subspace models (Everitt, Landau, Leese & Stahl, 2011).
// Factor Analysis and Multidimensional Scaling are used for visual representation of the interrelations among the variables by using quality data. This part will include the difference of cluster analysis from factor analysis and multidimensional scaling//. ...
The response addresses the queries posted in 1580 words with references.