International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 73 - Number 5 |
Year of Publication: 2013 |
Authors: Pallab Kumar Dey, Sripati Mukhopadhay |
10.5120/12734-9634 |
Pallab Kumar Dey, Sripati Mukhopadhay . Modified Deviation Approach to Deal with Missing Attribute Values in Data Mining with different Percentage of Missing Values. International Journal of Computer Applications. 73, 5 ( July 2013), 1-7. DOI=10.5120/12734-9634
Information System having missing attribute values (in practical) hampers accurate estimation of Data Mining. If missing attribute values can be predicted in the pre-processing stage of data mining then it will help to improve the accuracy, and the existing data mining algorithms can also be applied based on complete data. In this work different type of methods available to handle incomplete information system have been discussed, and there after an algorithm has been proposed by which missing attribute values may be replaced with minimum complexity. It is shown that proposed algorithm is better by applying it on different sets of data with different percentage of missing values.