Classical and Incremental Classification in Data Mining Process

Knowledge Discovery in Databases (KDD) is an iterative and multi step process that aims at extracting previously unknown and hidden patterns from a huge volume of databases. Data mining is a stage of the entire KDD process that involves applying a particular data mining algorithm to extract an interesting knowledge. One of the important problems that are used by data mining community is so-called classification problem. In this paper we study the classification task and provide a comprehensive study of classification techniques with more emphasis on classical and incremental decision tree based classification. While studying different classification techniques, we provide many important issues that distinguish between each classifier such as splitting criteria and pruning methods. Such criteria lead to the variation of decision tree based classification.