N-Similarity Binary Classifier (n-SBC)

A novel minimalist machine learning classifier based on Gray-coded binary patterns and Hamming similarity distance.

n-SBC (Similarity Binary Classifier)

A novel lazy/minimalist machine learning classifier based on Hamming binary similarity and Gray code (RBC).

How to Use it:

model = nsbc_train(X_train, y_train, num_decimals, u);
predictions = nsbc_predict(model, X_test);

Parameters:

• num_decimals: decimal places for normalization (e.g., 2)
• u: number of top neighbors to sum per class (e.g., 5)

Examples:

See examples:

• example_single_prediction.m: Example of train and predict only one sample using n-SBC model.
• example_loocv.m: Leave-One-Out Cross-Validation. For each sample, trains on all remaining samples and predicts the held-out one. Computes confusion matrix and balanced accuracy over the entire dataset.

New Python Package:

https://github.com/valdolab/n-sbc

https://pypi.org/project/nsbc/

References and based on the paper:

https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2025.1610856

Authors:

Osvaldo Velazquez-Gonzalez, Antonio Alarcón-Paredes and Cornelio Yañez-Marquez