Exploring Pauli Configurations in Quantum Kernels for Enhanced Binary Classification

This study pretends to set one’s sights on advancing the understanding of quantum machine learning’s applicability in binary tasks. By systematically assessing quantum kernels and their performance, the primary objective is to determine and evaluate how different Pauli configurations—precisely, ZZ, ZY, and a custom configuration (Z, YY, ZXZ)—impact execution accuracy running in simulator and 7-qubits IBM quantum hardware. The paper encompasses a robust methodology, including data preprocessing, quantum kernel creation, and integration into classical SVMs. It provides essential guidance for discussing the strengths and limitations of quantum methods, highlighting their potential as a valuable tool for developing accurate diagnostic models.