Data mining is the process in which useful information is extracted from large dataset. Such huge datasets consists of typically large number of patterns and features. This consumes a lot of storage space and since all of the data cannot be stored in main memory, it has to be fetched from secondary storage as required increasing the disk I/O operations. This situation can be resolved by using abstraction in data mining. Abstraction in simple terms refers to compact representation of dataset. Such an abstraction helps in reducing the time and space requirements of the overall decision making process.