Symbolic Data Analysis vs Classical Data Analysis: A Comparative Study

A symbolic data set is a combination of symbolic values. The analysis of these symbolic values is known as symbolic data analysis. It is an extension of the standard classical data analysis where symbolic data tables are used as input and symbolic objects are made output as a result. Symbolic data may arise in all branches of science and social science after aggregating a base data set over individual entries that together constitute a category of interest. This study attempts to bring into notice the use of symbolic data analysis and compare its outcome with standard classical data analysis. Different statistical tools have been used for comparative analysis of the symbolic and classical data viz. descriptive statistics, covariance, and correlation. To apply these statistical tools in both symbolic and classical data analysis set up, a well-known Iris flower data set is being used. The outcome of the study shows that there is a little difference in the results of descriptive statistics for the univariate case between classical data analysis and symbolic data analysis. However, in bivariate statistics computation though the directions of the covariance and correlation values (i.e. positive or negative) are the same, yet symbolic data analysis gives comparatively lesser magnitude values than the classical data analysis.