Efficient Approximate Solutions to Mutual Information Based Global Feature Selection
This work addresses feature selection for classifier models, offering incremental improvements in efficiency for a domain-specific task.
The paper tackles the intractable problem of mutual information-based feature selection by approximating it as a conditional mutual information problem, applying two global methods (TPower and LowRank) that show effectiveness across multiple datasets compared to existing procedures.
Mutual Information (MI) is often used for feature selection when developing classifier models. Estimating the MI for a subset of features is often intractable. We demonstrate, that under the assumptions of conditional independence, MI between a subset of features can be expressed as the Conditional Mutual Information (CMI) between pairs of features. But selecting features with the highest CMI turns out to be a hard combinatorial problem. In this work, we have applied two unique global methods, Truncated Power Method (TPower) and Low Rank Bilinear Approximation (LowRank), to solve the feature selection problem. These algorithms provide very good approximations to the NP-hard CMI based feature selection problem. We experimentally demonstrate the effectiveness of these procedures across multiple datasets and compare them with existing MI based global and iterative feature selection procedures.