Personalized Interpretable Classification

TL;DR

This paper introduces personalized interpretable classification, enabling models to generate individual-specific rules that enhance trustworthiness and potentially improve accuracy over existing methods.

Contribution

It formally defines the personalized interpretable classification problem and proposes greedy algorithms (PIC and fPIC) to generate individual-specific rules efficiently.

Findings

Personalized rules can identify test sample-specific insights missed by non-personalized classifiers.

Algorithms achieve comparable accuracy to state-of-the-art interpretable classifiers.

On breast cancer data, personalized classifiers outperform existing methods in accuracy and interpretability.

Abstract

How to interpret a data mining model has received much attention recently, because people may distrust a black-box predictive model if they do not understand how the model works. Hence, it will be trustworthy if a model can provide transparent illustrations on how to make the decision. Although many rule-based interpretable classification algorithms have been proposed, all these existing solutions cannot directly construct an interpretable model to provide personalized prediction for each individual test sample. In this paper, we make a first step towards formally introducing personalized interpretable classification as a new data mining problem to the literature. In addition to the problem formulation on this new issue, we present a greedy algorithm called PIC (Personalized Interpretable Classifier) to identify a personalized rule for each individual test sample. To improve the running efficiency, a fast approximate algorithm called fPIC is presented as well. To demonstrate the necessity, feasibility and advantages of such a personalized interpretable classification method, we conduct a series of empirical studies on real data sets. The experimental results show that: (1) The new problem formulation enables us to find interesting rules for test samples that may be missed by existing non-personalized classifiers. (2) Our algorithms can achieve the same-level predictive accuracy as those state-of-the-art (SOTA) interpretable classifiers. (3) On a real data set for predicting breast cancer metastasis, such personalized interpretable classifiers can outperform SOTA methods in terms of both accuracy and interpretability.