An Analysis of Data Characteristics and Classifier Performance in the Context of Ordinal Classification

The characteristics of data set play an important and foremost role in data mining and knowledge engineering. Classification is a foremost and necessary method in data mining. The motto of classification is to investigate the training data and to acquire an accurate description or model for each class using the characteristics present in the data. The characteristics of the data set are one of the leading performance affecting factors of the classifier. This paper performs an analysis of data set characteristics and its correlation with classifier performance. In particular, this analysis extends the study to the issues of ordinal classification with respect to the data set and recommendations proposed to address these issues.