Predicting Missing Values in a Dataset: Challenges and Approaches.

Data serves as an asset in the present era of information in almost all fields of study. The quality of the knowledge extracted, learning and decision problems depends upon the quality of data gathered. But most of the real world datasets tends to be incomplete. The problem of missing data can have a significant effect on the conclusions that can be drawn from this data. Data mining has made a great progress in recent year but the problem of missing data or value has still remained a great challenge for data mining. Missing data or value in a dataset can affect the performance of classifier which leads to difficulty of extracting useful information from datasets. In this paper we present the major challenges in predicting missing data in a data set and the existing approaches dealing with the missing data imputation. [ABSTRACT FROM AUTHOR]