An Expected Utility Approach to Active Feature-Value Acquisition

Maytal Saar-Tsechansky, Prem Melville, Foster J. Provost, Raymond J. Mooney

Abstract

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.
BibTeX
@inproceedings{saartsechansky2005expected,
  author = {Maytal Saar-Tsechansky and Prem Melville and Foster J. Provost and Raymond J. Mooney},
  title = {{An Expected Utility Approach to Active Feature-Value Acquisition}},
  booktitle = {ICDM},
  year = {2005},
  doi = {10.1109/icdm.2005.23},
}

Loading related content...

Figure