A Study on different data mining classifiers

Abstract

Data mining is a process of finding patterns in a large dataset. It involves various algorithmic classifiers. Almost everyone in the IT sector is utilizing data mining techniques to understand different patterns in a large dataset. Classification is required to find out in which group the given instance of the testing dataset is related to a given class of a training dataset. It partitions given information into its subclasses which are dependent upon some observed parameters. Several different types of techniques used for data mining are - Decision Trees (ID3, C4.5, CART), k-nearest neighbours, Apriori algorithm, Naive Bayes, Neural Networks. Comprehensively, The three approaches followed for classification technique are Machine Learning, Statistical Based and Neural Networks. Considering these approaches broadly we can define different classifiers. The classification has innumerable applications in speech recognition, computer vision, Geostatistics, Biological classification. This study presents a comprehensive survey of different data mining classifiers and compares them on different parameters.