Cargando…
Mixed Quantum–Classical Method for Fraud Detection With Quantum Feature Selection
This article presents a first end-to-end application of a quantum support vector machine (QSVM) algorithm for a classification problem in the financial payment industry using the IBM Safer Payments and IBM Quantum Computers via the Qiskit software stack. Based on real card payment data, a thorough c...
Autores principales: | , , , , , , |
---|---|
Lenguaje: | eng |
Publicado: |
2022
|
Materias: | |
Acceso en línea: | https://dx.doi.org/10.1109/TQE.2022.3213474 http://cds.cern.ch/record/2843768 |
_version_ | 1780976319410995200 |
---|---|
author | Grossi, Michele Ibrahim, Noelle Radescu, Voica Loredo, Robert Voigt, Kirsten Von Altrock, Constantin Rudnik, Andreas |
author_facet | Grossi, Michele Ibrahim, Noelle Radescu, Voica Loredo, Robert Voigt, Kirsten Von Altrock, Constantin Rudnik, Andreas |
author_sort | Grossi, Michele |
collection | CERN |
description | This article presents a first end-to-end application of a quantum support vector machine (QSVM) algorithm for a classification problem in the financial payment industry using the IBM Safer Payments and IBM Quantum Computers via the Qiskit software stack. Based on real card payment data, a thorough comparison is performed to assess the complementary impact brought in by the current state-of-the-art quantum machine-learning algorithms with respect to the classical approach. A new method to search for best features is explored using the QSVM's feature map characteristics. The results are compared using fraud-specific key performance indicators, i.e., accuracy, recall, and false positive rate, extracted from analyses based on human expertise (such as rule decisions), classical machine-learning algorithms (such as random forest and XGBoost), and quantum-based machine-learning algorithms using QSVM. In addition, a hybrid classical–quantum approach is explored by using an ensemble model that combines classical and quantum algorithms to better improve the fraud prevention decision. We found, as expected, that the results highly depend on feature selections and algorithms that are used to select them. The QSVM provides a complementary exploration of the feature space that led to an improved accuracy of the mixed quantum-classical method for fraud detection, on a drastically reduced dataset to fit current state of quantum hardware. |
id | cern-2843768 |
institution | Organización Europea para la Investigación Nuclear |
language | eng |
publishDate | 2022 |
record_format | invenio |
spelling | cern-28437682023-09-29T02:27:17Zdoi:10.1109/TQE.2022.3213474http://cds.cern.ch/record/2843768engGrossi, MicheleIbrahim, NoelleRadescu, VoicaLoredo, RobertVoigt, KirstenVon Altrock, ConstantinRudnik, AndreasMixed Quantum–Classical Method for Fraud Detection With Quantum Feature Selectionquant-phcs.LGQuantum TechnologyComputing and ComputersThis article presents a first end-to-end application of a quantum support vector machine (QSVM) algorithm for a classification problem in the financial payment industry using the IBM Safer Payments and IBM Quantum Computers via the Qiskit software stack. Based on real card payment data, a thorough comparison is performed to assess the complementary impact brought in by the current state-of-the-art quantum machine-learning algorithms with respect to the classical approach. A new method to search for best features is explored using the QSVM's feature map characteristics. The results are compared using fraud-specific key performance indicators, i.e., accuracy, recall, and false positive rate, extracted from analyses based on human expertise (such as rule decisions), classical machine-learning algorithms (such as random forest and XGBoost), and quantum-based machine-learning algorithms using QSVM. In addition, a hybrid classical–quantum approach is explored by using an ensemble model that combines classical and quantum algorithms to better improve the fraud prevention decision. We found, as expected, that the results highly depend on feature selections and algorithms that are used to select them. The QSVM provides a complementary exploration of the feature space that led to an improved accuracy of the mixed quantum-classical method for fraud detection, on a drastically reduced dataset to fit current state of quantum hardware.This paper presents a first end-to-end application of a Quantum Support Vector Machine (QSVM) algorithm for a classification problem in the financial payment industry using the IBM Safer Payments and IBM Quantum Computers via the Qiskit software stack. Based on real card payment data, a thorough comparison is performed to assess the complementary impact brought in by the current state-of-the-art Quantum Machine Learning algorithms with respect to the Classical Approach. A new method to search for best features is explored using the Quantum Support Vector Machine's feature map characteristics. The results are compared using fraud specific key performance indicators: Accuracy, Recall, and False Positive Rate, extracted from analyses based on human expertise (rule decisions), classical machine learning algorithms (Random Forest, XGBoost) and quantum based machine learning algorithms using QSVM. In addition, a hybrid classical-quantum approach is explored by using an ensemble model that combines classical and quantum algorithms to better improve the fraud prevention decision. We found, as expected, that the results highly depend on feature selections and algorithms that are used to select them. The QSVM provides a complementary exploration of the feature space which led to an improved accuracy of the mixed quantum-classical method for fraud detection, on a drastically reduced data set to fit current state of Quantum Hardware.arXiv:2208.07963oai:cds.cern.ch:28437682022 |
spellingShingle | quant-ph cs.LG Quantum Technology Computing and Computers Grossi, Michele Ibrahim, Noelle Radescu, Voica Loredo, Robert Voigt, Kirsten Von Altrock, Constantin Rudnik, Andreas Mixed Quantum–Classical Method for Fraud Detection With Quantum Feature Selection |
title | Mixed Quantum–Classical Method for Fraud Detection With Quantum Feature Selection |
title_full | Mixed Quantum–Classical Method for Fraud Detection With Quantum Feature Selection |
title_fullStr | Mixed Quantum–Classical Method for Fraud Detection With Quantum Feature Selection |
title_full_unstemmed | Mixed Quantum–Classical Method for Fraud Detection With Quantum Feature Selection |
title_short | Mixed Quantum–Classical Method for Fraud Detection With Quantum Feature Selection |
title_sort | mixed quantum–classical method for fraud detection with quantum feature selection |
topic | quant-ph cs.LG Quantum Technology Computing and Computers |
url | https://dx.doi.org/10.1109/TQE.2022.3213474 http://cds.cern.ch/record/2843768 |
work_keys_str_mv | AT grossimichele mixedquantumclassicalmethodforfrauddetectionwithquantumfeatureselection AT ibrahimnoelle mixedquantumclassicalmethodforfrauddetectionwithquantumfeatureselection AT radescuvoica mixedquantumclassicalmethodforfrauddetectionwithquantumfeatureselection AT loredorobert mixedquantumclassicalmethodforfrauddetectionwithquantumfeatureselection AT voigtkirsten mixedquantumclassicalmethodforfrauddetectionwithquantumfeatureselection AT vonaltrockconstantin mixedquantumclassicalmethodforfrauddetectionwithquantumfeatureselection AT rudnikandreas mixedquantumclassicalmethodforfrauddetectionwithquantumfeatureselection |