Knowledge Integrated Classifier Design Based on Utility Optimization

TL;DR

This paper introduces a framework for designing classifiers that incorporate prior knowledge to optimize a utility function, ensuring asymptotic convergence to an extended Bayes rule as data size increases.

Contribution

It presents a systematic method for integrating domain knowledge into classifier design with theoretical guarantees of convergence to an optimal utility-maximizing classifier.

Findings

Classifier asymptotically converges to the optimal utility-based classifier

Provides a theoretical foundation for knowledge-guided classification

Enables domain experts to influence classification priorities

Abstract

This paper proposes a systematic framework to design a classification model that yields a classifier which optimizes a utility function based on prior knowledge. Specifically, as the data size grows, we prove that the produced classifier asymptotically converges to the optimal classifier, an extended version of the Bayes rule, which maximizes the utility function. Therefore, we provide a meaningful theoretical interpretation for modeling with the knowledge incorporated. Our knowledge incorporation method allows domain experts to guide the classifier towards correctly classifying data that they think to be more significant.