More Conference Papers
Profit-Optimal Nitrogen Under Climate Change: DSSAT-Derived EONR Across Global Staple Systems
J. ET-TOUSY, S. ET-TOUSY, Y. MAMMA, A. ZYANE
AIST4GE'26 — The 1st International Conference on Artificial Intelligence and Smart Technologies for Green Energy
Saidia, Morocco
This study develops a high-throughput simulation-economic framework that converts DSSAT v4.8 from a field-scale crop model into a scalable decision tool for nitrogen management under climate change. It quantifies how climate forcing alters the nitrogen yield-response of maize, rice, wheat, and soybean across major producing countries and agro-ecological zones under CMIP6 SSP pathways and evaluates economically optimal nitrogen rates using the Economically Optimal Nitrogen Rate (EONR).
Lightweight Machine Learning Framework for QoS-Aware oneM2M Deployments at the Edge
J. Et-Tousy, A. Zyane
FiCloud 2026 — 13th IEEE International Conference on Future Internet of Things and Cloud
Granada, Spain
This work proposes a lightweight machine-learning framework for QoS-aware oneM2M deployments at the edge. It uses resource-efficient models to classify service conditions and support timely QoS decisions under constrained computing and networking environments.
Fuzzy-Logic-Based QoS Control in oneM2M: Adaptive Edge-to-Cloud Traffic Offloading
A. Zyane, J. Et-Tousy
FiCloud 2026 — 13th IEEE International Conference on Future Internet of Things and Cloud
Granada, Spain
This paper presents a fuzzy-logic controller for adaptive traffic offloading between edge and cloud resources in oneM2M systems. The controller considers QoS and resource indicators to improve latency, service continuity, and infrastructure utilization.
Comparative QoS Analysis Between ITU-T Requirements and an Enhanced oneM2M Middleware: A Proof of Concept
J. Et-Tousy, A. Zyane
AIST4GE 2026 — 1st International Conference on Artificial Intelligence and Smart Technologies for Green Energy
Saidia, Morocco
This proof of concept compares ITU-T quality-of-service requirements with the behavior of an enhanced oneM2M middleware. The evaluation examines communication protocols and system indicators to determine how effectively the middleware satisfies application-level QoS constraints.
Balancing Security and QoS in oneM2M IoT Systems: A Multi-Objective Machine Learning Framework
J. Et-Tousy, A. Zyane, H. Jamiri
ISDA 2026 — International Conference on Intelligent Systems and Digital Applications
Settat, Morocco
This study introduces a multi-objective machine-learning framework that jointly considers security effectiveness and QoS performance in oneM2M IoT systems. It analyzes attack detection alongside latency, reliability, and resource consumption to support balanced operational decisions.
A Modular Multi-Agent Multimodal AI Framework Integrating Vision-Language Models and IoT Data for Explainable Clinical Decision Support
J. Et-Tousy, M. Chtara, A. Zyane
IRASET 2026 — 6th International Conference on Innovative Research in Applied Science, Engineering and Technology
Fez, Morocco
This paper proposes a modular multi-agent framework that combines vision-language models with IoT-generated clinical data. The architecture coordinates specialized agents to produce explainable and context-aware decision support from heterogeneous medical information.
Autonomic QoS Management in oneM2M for IoT Using AI Agent
J. Et-Tousy, A. Zyane
COCIA 2026 — 4th International Conference on Connected Objects and Artificial Intelligence
Casablanca, Morocco
This work presents an AI-agent-based approach for autonomic QoS management in oneM2M IoT environments. The agent monitors system indicators, analyzes service degradation, and selects adaptive actions to maintain latency, reliability, and resource-usage objectives.
Black-Box Adversarial Threats in oneM2M-Based IoT Networks: Evaluating and Defending Intrusion Detection Systems
H. Jamiri, A. Zyane, J. Et-Tousy
COCIA 2026 — 4th International Conference on Connected Objects and Artificial Intelligence
Casablanca, Morocco
This study evaluates the vulnerability of intrusion-detection systems to black-box adversarial attacks in oneM2M-based IoT networks. It examines attack impact and defensive strategies intended to improve the robustness of machine-learning security models.
Towards Adaptive Learning in Literature Education: A Structured Data Approach Using NLP Techniques
J. Et-Tousy, A. Zyane
ADACIS 2025 — 2nd IEEE International Conference on Advances in Data-Driven Analytics and Intelligent Systems
Sousse, Tunisia
This paper presents a structured-data approach for adaptive learning in literature education using natural-language-processing techniques. It organizes educational content and learner-related information to support personalized activities, assessment, and pedagogical recommendations.
Machine Learning-Driven QoS Optimization for IoT in oneM2M
J. Et-Tousy, A. Zyane
FiCloud 2025 — 12th International Conference on Future Internet of Things and Cloud
Istanbul, Türkiye
This paper applies machine-learning techniques to optimize QoS in oneM2M-based IoT systems. Models use latency, resource consumption, success rate, and throughput indicators to classify operating conditions and support adaptive service management.