A comparative and practical approach using quantum machine learning (QML) and support vector classifier (SVC) for Light emitting diodes mediated in vitro micropropagation of black mulberry (Morus nigra L.)

Abstract

In this study, in vitro regeneration protocol for black mulberry (Morus nigra L.) was optimized using 18 distinct combinations of benzylaminopurine (BAP) with either naphthalene acetic acid (NAA) or Indole butyric acid (IBA). The top two combinations were then utilized to optimize the light intensity given by light-emitting diodes (LEDs). Supplementation of 0.5 mg L-1 BAP and 0.25 mg L-1 IBA with 60 PPFD light intensity yielded a maximum regeneration coefficient (2.53), shoot length (6.01 cm), and number of leaves (10.73). The regenerated plantlets were rooted with IBA under in vitro conditions followed by successful acclimatization of plantlets under greenhouse conditions. The results were further investigated by linking them with an emphasis on improving the Support Vector Classifier (SVC) using quantum computing techniques, and this work embarked on a groundbreaking path to integrate the realms of machine learning (ML) with quantum computing. For this purpose, the traditional Support Vector Classifier (SVC) model was compared with quantum-enhanced algorithms, including SVC with the quantum kernel (SVC Qkernel), SVC with quantum features (SVC Qfeatures), Quantum Support Vector Classifier (QSVC), and the Variational Quantum Circuit (VQC). The quantum-enhanced models showed a range of results, indicating their complex and subtle character, whereas classical SVC performed robustly for multiple metrics. Quantum kernel-based SVC demonstrated an interesting trade-off between recall and precision, indicating its proficiency in processing particular data properties.