Abstract:In machine learning applications, the reliability of predictions is significant for assisted decision and risk control. As an effective framework to quantify the prediction reliability, conformal prediction (CP) was developed with the CPKNN (CP with kNN). However, the conventional CPKNN suffers from high variance and bias and long computational time as the feature dimensionality increases. To address these limitations, a new CP framework-conformal prediction with shrunken centroids (CPSC) is proposed. It regularizes the class centroids to attenuate the irrelevant features and shrink the sample space for predictions and reliability quantification. To compare CPKNN and CPSC, we employed them in the classification of 12 categories of alternative herbal medicine with electronic nose as a case and assessed them in two tasks: 1) offline prediction: the training set was fixed and the accuracy on the testing set was evaluated; 2) online prediction with data augmentation: they filtered unlabeled data to augment the training data based on the prediction reliability and the final accuracy of testing set was compared. The result shows that CPSC significantly outperformed CPKNN in both two tasks: 1) CPSC reached a significantly higher accuracy with lower computation cost, and with the same credibility output, CPSC generally achieves a higher accuracy; 2) the data augmentation process with CPSC robustly manifested a statistically significant improvement in prediction accuracy with different reliability thresholds, and the augmented data were more balanced in classes. This novel CPSC provides higher prediction accuracy and better reliability quantification, which can be a reliable assistance in decision support.

Assessment of Conformal Prediction and Standard Normal Distribution for Autonomous Consensus One‐Class Classification

Gaining Confidence on Molecular Classification Through Consensus Modeling and Validation

Conformal predictions for electronic nose system: An application to tea classification

Posterior Conformal Prediction

Confidence on the Focal: Conformal Prediction with Selection-Conditional Coverage

Class-Conditional Conformal Prediction with Many Classes

conformalClassification: A Conformal Prediction R Package for Classification

Trustworthy Classification through Rank-Based Conformal Prediction Sets

Conformal Prediction via Regression-as-Classification

Conformal Classification with Equalized Coverage for Adaptively Selected Groups

Conformal Risk Control for Ordinal Classification

Out-of-Distribution Detection Should Use Conformal Prediction (and Vice-versa?)

CPSC: Conformal prediction with shrunken centroids for efficient prediction reliability quantification and data augmentation, a case in alternative herbal medicine classification with electronic nose

Nothing Conformal about Adaptive Conformal Inference

Spatial-Aware Conformal Prediction for Trustworthy Hyperspectral Image Classification

The Penalized Inverse Probability Measure for Conformal Classification

On Temperature Scaling and Conformal Prediction of Deep Classifiers

Beyond Conformal Predictors: Adaptive Conformal Inference with Confidence Predictors

Classification with Valid and Adaptive Coverage

Nested conformal prediction and quantile out-of-bag ensemble methods

Conjunction Subspaces Test for Conformal and Selective Classification