AI-Driven F2N-ECC Framework for Secure, Real-Time E-Commerce Transactions

Fatma Mallouli; Rimah Amami; Eman Jaradat; Hassan Ahmad Eleraky; Ahmad Abas; Rim Amami; Najla Frih

doi:10.1016/j.procs.2025.09.591

AI-Driven F2N-ECC Framework for Secure, Real-Time E-Commerce Transactions

Fatma Mallouli^*
, Rimah Amami
, Eman Jaradat
, Hassan Ahmad Eleraky
, Ahmad Abas
, Rim Amami
, Najla Frih

^*Corresponding author for this work

Imam Abdulrahman Bin Faisal University
Department of Basic Sciences (Mathematics and Statistics)

Research output: Contribution to journal › Conference article › peer-review

Abstract

As e-commerce continues to expand rapidly, ensuring secure real-time transactions has become increasingly important. This paper introduces an innovative cryptographic framework powered by artificial intelligence, integrating a Long Short-Term Memory (LSTM) based prediction engine with a Field-to-Number enhanced Elliptic Curve Cryptography (F2N-ECC) module. Designed for adaptive and secure processing, the system dynamically adjusts encryption strength based on real-time fraud risk assessments, while blockchain-based logging ensures tamper-proof transaction verification. Tests using a labeled e-commerce dataset show promising results: 96.4% accuracy in fraud detection, a 50% reduction in latency thanks to F2N-ECC, and strong scalability for Internet of Things (IoT) environments. Additionally, a case study on anomaly detection in financial time-series data highlights the framework's flexibility across different domains. Overall, this architecture paves the way for AI-driven, post-quantum resilient security solutions tailored for the future of digital commerce.

Original language	English
Pages (from-to)	4655-4666
Number of pages	12
Journal	Procedia Computer Science
Volume	270
DOIs	https://doi.org/10.1016/j.procs.2025.09.591
State	Published - 2025
Event	29th International Conference on Knowledge-Based and Intelligent Information and Engineering Systems, KES 2025 - Osaka, Japan Duration: 10 Sep 2025 → 12 Sep 2025

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 8 Decent Work and Economic Growth
SDG 9 Industry, Innovation, and Infrastructure

Keywords

AI-driven security
Blockchain
E-Commerce
Elliptic Curve Cryptography
Field-to-Number Transformation
Fraud Detection
Post-Quantum Cryptography
Sensor Networks

Access to Document

10.1016/j.procs.2025.09.591

Cite this

@article{15d502c6e21a40fb9092f74b3e86f8d3,

title = "AI-Driven F2N-ECC Framework for Secure, Real-Time E-Commerce Transactions",

abstract = "As e-commerce continues to expand rapidly, ensuring secure real-time transactions has become increasingly important. This paper introduces an innovative cryptographic framework powered by artificial intelligence, integrating a Long Short-Term Memory (LSTM) based prediction engine with a Field-to-Number enhanced Elliptic Curve Cryptography (F2N-ECC) module. Designed for adaptive and secure processing, the system dynamically adjusts encryption strength based on real-time fraud risk assessments, while blockchain-based logging ensures tamper-proof transaction verification. Tests using a labeled e-commerce dataset show promising results: 96.4\% accuracy in fraud detection, a 50\% reduction in latency thanks to F2N-ECC, and strong scalability for Internet of Things (IoT) environments. Additionally, a case study on anomaly detection in financial time-series data highlights the framework's flexibility across different domains. Overall, this architecture paves the way for AI-driven, post-quantum resilient security solutions tailored for the future of digital commerce.",

keywords = "AI-driven security, Blockchain, E-Commerce, Elliptic Curve Cryptography, Field-to-Number Transformation, Fraud Detection, Post-Quantum Cryptography, Sensor Networks",

author = "Fatma Mallouli and Rimah Amami and Eman Jaradat and Eleraky, \{Hassan Ahmad\} and Ahmad Abas and Rim Amami and Najla Frih",

note = "Publisher Copyright: {\textcopyright} 2025 The Authors. Published by Elsevier B.V.; 29th International Conference on Knowledge-Based and Intelligent Information and Engineering Systems, KES 2025 ; Conference date: 10-09-2025 Through 12-09-2025",

year = "2025",

doi = "10.1016/j.procs.2025.09.591",

language = "English",

volume = "270",

pages = "4655--4666",

journal = "Procedia Computer Science",

issn = "1877-0509",

}

TY - JOUR

T1 - AI-Driven F2N-ECC Framework for Secure, Real-Time E-Commerce Transactions

AU - Mallouli, Fatma

AU - Amami, Rimah

AU - Jaradat, Eman

AU - Eleraky, Hassan Ahmad

AU - Abas, Ahmad

AU - Amami, Rim

AU - Frih, Najla

PY - 2025

Y1 - 2025

N2 - As e-commerce continues to expand rapidly, ensuring secure real-time transactions has become increasingly important. This paper introduces an innovative cryptographic framework powered by artificial intelligence, integrating a Long Short-Term Memory (LSTM) based prediction engine with a Field-to-Number enhanced Elliptic Curve Cryptography (F2N-ECC) module. Designed for adaptive and secure processing, the system dynamically adjusts encryption strength based on real-time fraud risk assessments, while blockchain-based logging ensures tamper-proof transaction verification. Tests using a labeled e-commerce dataset show promising results: 96.4% accuracy in fraud detection, a 50% reduction in latency thanks to F2N-ECC, and strong scalability for Internet of Things (IoT) environments. Additionally, a case study on anomaly detection in financial time-series data highlights the framework's flexibility across different domains. Overall, this architecture paves the way for AI-driven, post-quantum resilient security solutions tailored for the future of digital commerce.

AB - As e-commerce continues to expand rapidly, ensuring secure real-time transactions has become increasingly important. This paper introduces an innovative cryptographic framework powered by artificial intelligence, integrating a Long Short-Term Memory (LSTM) based prediction engine with a Field-to-Number enhanced Elliptic Curve Cryptography (F2N-ECC) module. Designed for adaptive and secure processing, the system dynamically adjusts encryption strength based on real-time fraud risk assessments, while blockchain-based logging ensures tamper-proof transaction verification. Tests using a labeled e-commerce dataset show promising results: 96.4% accuracy in fraud detection, a 50% reduction in latency thanks to F2N-ECC, and strong scalability for Internet of Things (IoT) environments. Additionally, a case study on anomaly detection in financial time-series data highlights the framework's flexibility across different domains. Overall, this architecture paves the way for AI-driven, post-quantum resilient security solutions tailored for the future of digital commerce.

KW - AI-driven security

KW - Blockchain

KW - E-Commerce

KW - Elliptic Curve Cryptography

KW - Field-to-Number Transformation

KW - Fraud Detection

KW - Post-Quantum Cryptography

KW - Sensor Networks

UR - https://www.scopus.com/pages/publications/105024078887

U2 - 10.1016/j.procs.2025.09.591

DO - 10.1016/j.procs.2025.09.591

M3 - Conference article

AN - SCOPUS:105024078887

SN - 1877-0509

VL - 270

SP - 4655

EP - 4666

JO - Procedia Computer Science

JF - Procedia Computer Science

T2 - 29th International Conference on Knowledge-Based and Intelligent Information and Engineering Systems, KES 2025

Y2 - 10 September 2025 through 12 September 2025

ER -

AI-Driven F2N-ECC Framework for Secure, Real-Time E-Commerce Transactions

Abstract

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this