An Expected Utility Approach to Active Feature-Value Acquisition

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 costeffective 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.

• Type: Conference
• Year: 2005
• Where: ICDM
• Authors: Prem Melville, Foster J. Provost, Raymond J. Mooney
• Keywords: Data Mining, Feature Values, ICDM 2005, Missing Values, Model’s Accuracy