Data Pre-processing Techniques in Data Mining: A Review

Abstract

Data mining is the process of finding  interesting patterns and models from massive datasets. In  the field of natural and physical sciences, data collection,  management, and analysis have evolved as the most  trustworthy source of information and emergence of new  findings, information, and products. The development of  the most effective procedures in statistical circumstances  has therefore become standard practice in the academic and industry sectors. Under actual situations, dealing with  enormous datasets, there are bound to be discrepancies  and abnormalities of many types that prohibit us from  knowing the true results of realistic issues. These  concepts and trends are helpful in decision-making  situations. The quality of the data is the most important  factor in data mining. For efficient information mining,  computer-based data pre-processing approaches provide  methods that assist the data under processing in  conforming to conventional structures, hence  significantly improving the efficiency of  computer algorithms.