In many classification tasks training data have missing feature values that can be acquired at a cost. For building accurate predictive models, acquiring all missing values is often prohibitively expensive or unnecessary, while acquiring a random subset of feature values may not be most effective. The goal of active feature-value acquisition is to incrementally select feature values that are most cost-effective for improving the model's accuracy. We present an approach that acquires feature values for inducing a classification model based on an estimation of the expected improvement in model accuracy per unit cost. Experimental results demonstrate that our approach consistently reduces the cost of producing a model of a desired accuracy compared to random feature acquisitions.

PDF, PS

In Proceedings of the International Conference on Data Mining, pp. 745-748, Houston, TX, November 2005.

@InProceedings{melville:icdm05, title={An Expected Utility Approach to Active Feature-value Acquisition}, author={P. Melville and M. Saar-Tsechansky and F. Provost and Raymond J. Mooney}, booktitle={Proceedings of the International Conference on Data Mining}, month={November}, address={Houston, TX}, pages={745-748}, url="http://www.cs.utexas.edu/users/ai-lab?melville:icdm05", year={2005} }