Handling missing attribute values in preterm birth data sets

Abstract

The objective of our research was to find the best approach to handle missing attribute values in data sets describing preterm birth provided by the Duke University. Five strategies were used for filling in missing attribute values, based on most common values and closest fit for symbolic attributes, averages for numerical attributes, and a special approach to induce only certain rules from specified information using the MLEM2 approach. The final conclusion is that the best strategy was to use the global most common method for symbolic attributes and the global average method for numerical attributes. © Springer-Verlag Berlin Heidelberg 2005.