Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Aristeidis Mystakidis has a total of 51 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Short-Term Traffic Speed Prediction and Pattern Discovery Using Sensor Data

    by Ali Hadi, Muhammad Munim Shafi, Asadullah Safi, Aristeidis Mystakidis, Christos Tjortjis

    The sustained growth of urban populations has worsened chronic traffic congestion, with detrimental impacts on multiple dimensions of urban livability, including increased commute times, reduced road safety, and the degradation of local air quality. Thus, traffic speed prediction (TSP) is an important component of intelligent transportation systems (ITSs) because urban traffic behavior is dynamic, nonlinear, and time-dependent. This study addresses the challenge of short-term TSP using time-series regression techniques applied to the METR-LA dataset utilizing a multi-sensor subset containing one week of data. After reconstruction into a tabular format, the data were preprocessed using interpolation via forward/backward filling, and transformed using temporal and lag-based features. On the modeling side, a comparative analysis with Linear Regression (LR), K-Nearest Neighbors (KNN), Decision Tree (DT), and Random Forest (RF) models was conducted, while clustering provided additional insights into traffic-state patterns. Results indicated that KNN achieved the best performance after tuning (MAE = 2.22, RMSE = 4.38, R2 = 0.899). The findings illustrated that simple interpretable models can support TSP and traffic-state understanding.

  • Open Access

    Operations research: a comprehensive literature review on nurse rostering optimisation

    by Paraskevas Koukaras, Christos Koukaras, Aristeidis Mystakidis, Georgios D. Mantzavinis, Stavros G. Stavrinides, Christos Tjortjis

    This comprehensive review aims to address a complex part of operations research, the nurse rostering problem. It is a critical piece of healthcare management, which affects the operational efficiency of hospitals and, at the same time, the quality of patient care. Given the dynamics of the diversity in the healthcare settings and needs, optimising the nurse staff schedule not only improves the care delivery, but also affects staff satisfaction. This review covers a wide range of techniques, including heuristics, meta-heuristics, such as simulated annealing, variable neighborhood search, genetic algorithms, population-based meta-heuristics, such as plant propagation algorithm, bee colony optimisation, and differential evolution, and advanced approaches, like hyper-heuristics, stochastic programming, and hybrid approaches. Additionally, the study explores mathematical optimisation techniques, including integer programming, mixed integer programming, and branch-and-price algorithms, with a special focus on hybrid methods, which combine meta-heuristics with exact optimisation approaches. By analysing key and recent studies, the review shows that hybrid approaches, particularly those integrating meta-heuristics with mathematical models, appear to be the most effective in addressing the complexities of nurse rostering (NR). Future directions, on the other hand, indicate that the growing integration of machine learning and real-time scheduling systems can provide further adaptability and scalability for NR solutions.


  • Open Access

    XAI-Driven Malware Detection from Memory Artifacts: An Alert-Driven AI Framework with TabNet and Ensemble Classification

    by Aristeidis Mystakidis, Grigorios Kalogiannnis, Nikolaos Vakakis, Nikolaos Altanis, Konstantina Milousi, Iason Somarakis, Gabriela Mihalachi, Mariana S. Mazi, Dimitris Sotos, Antonis Voulgaridis, et al.

    Modern malware presents significant challenges to traditional detection methods, often leveraging fileless techniques, in-memory execution, and process injection to evade antivirus and signature-based systems. To address these challenges, alert-driven memory forensics has emerged as a critical capability for uncovering stealthy, persistent, and zero-day threats. This study presents a two-stage host-based malware detection framework, that integrates memory forensics, explainable machine learning, and ensemble classification, designed as a post-alert asynchronous SOC workflow balancing forensic depth and operational efficiency. Utilizing the MemMal-D2024 dataset—comprising rich memory forensic artifacts from Windows systems infected with malware samples whose creation metadata spans 2006–2021—the system performs malware detection, using features extracted from volatile memory. In the first stage, an Attentive and Interpretable Learning for structured Tabular data (TabNet) model is used for binary classification (benign vs. malware), leveraging its sequential attention mechanism and built-in explainability. In the second stage, a Voting Classifier ensemble, composed of Light Gradient Boosting Machine (LGBM), eXtreme Gradient Boosting (XGB), and Histogram Gradient Boosting (HGB) models, is used to identify the specific malware family (Trojan, Ransomware, Spyware). To reduce memory dump extraction and analysis time without compromising detection performance, only a curated subset of 24 memory features—operationally selected to reduce acquisition/extraction time and validated via redundancy inspection, model explainability (SHAP/TabNet), and training data correlation analysis —was used during training and runtime, identifying the best trade-off between memory analysis and detection accuracy. The pipeline, which is triggered from host-based Wazuh Security Information and Event Management (SIEM) alerts, achieved 99.97% accuracy in binary detection and 70.17% multiclass accuracy, resulting in an overall performance of 87.02%, including both global and local explainability, ensuring operational transparency and forensic interpretability. This approach provides an efficient and interpretable detection solution used in combination with conventional security tools as an extra layer of defense suitable for modern threat landscapes.


  • Open Access

    EV charging forecasting exploiting traffic, weather and user information

    by Aristeidis Mystakidis, Nikolaos Tsalikidis, Paraskevas Koukaras, Georgios Skaltsis, Dimosthenis Ioannidis, Christos Tjortjis, Dimitrios Tzovaras

    Driven by the urgency of transitioning to a decarbonised energy sector, the confluence of Renewable Energy Sources (RES), Smart Grids (SG), and the widespread adoption of Electric Vehicles (EVs) has made power system planning and operation a focal point of global sustainability efforts. The rapid adoption of EVs has led to challenges related to charging infrastructure, grid demand, and scheduling management while also offering opportunities as a distributed energy storage solution, aiding peak load reduction and renewable energy integration. Enhanced accuracy in EV charging predictions could alleviate energy imbalances resulting from production and consumption disparities. Additionally, it supports SG architectural functions like demand response (DR) management. Compared to the initial baseline timeseries EV charging forecasting models, which performed poorly, the proposed strategy consists of a step-by-step forecasting methodology using data from predictions of road traffic, weather/seasonality, user charging information, and EV charging load. For each target, various novel implementations and models were conducted, including zero inflated (ZI) and a novel logarithmic ZI (LogZI) regression method. Results comparison indicated an increasing accuracy of EV charging demand forecasts from 78.1 to 88.7% regarding $$R^2$$ R 2 . It also highlighted a decreasing error from 1.75 to 1.05 kWh regarding MAE and from 2.48 to 1.77 kWh regarding RMSE (CVRMSE and NRMSE were also provided). This research contributes a novel one step ahead EV charging demand forecasting framework integrating traffic and user charging predictions with the proposed LogZI regression approach. The proposed, updated version of the ZI approach, methodology performs significantly better compared to the baseline models, aiming to enhance grid stability and optimize charging infrastructure.


  • Open Access

    Continual Error Correction on Low-Resource Devices

    by Kirill Paramonov, Mete Ozay, Aristeidis Mystakidis, Nikolaos Tsalikidis, Dimitrios Sotos, Anastasios Drosou, Dimitrios Tzovaras, Hyunjun Kim, Kiseok Chang, Sangdok Mo, et al.


  • Open Access

    A multi-energy meta-model strategy for multi-step ahead energy load forecasting

    by Aristeidis Mystakidis, Evangelia Ntozi, Paraskevas Koukaras, Nikolaos Katsaros, Dimosthenis Ioannidis, Christos Tjortjis, Dimitrios Tzovaras

    Energy load forecasting (ELF) is necessary for optimal scheduling of electricity distribution to customers, reducing energy losses and for regulating power distribution. There are numerous approaches for ELF based on statistical, Machine Learning (ML) and deep learning (DL) methods. This paper performs a comparative multi-step ahead ELF utilising rolling (or recursive) versus direct forecasting techniques for three energy types (ETs), i.e. heating, cooling and electricity. It introduces a multi-energy meta-model strategy (MEMMS), a unified approach that combines the best outcomes of the comparison and utilises one model per time step ahead for all ETs. The findings were validated on three datasets, one per ET, employing metrics such as mean absolute error (MAE), root-mean-squared error (RMSE), $$R^{2}$$ R 2 and execution time. The novel aspects of this work can be attributed to the possible identification of the best-performing ML and DL models across all ETs, as well as the multi-energy behaviour recognition of the multi-step ahead strategies and the employed MEMMS approach. Initial comparison results indicated that direct multi-step ahead voting ensemble with light gradient boosting machine (LGBM) and CatBoost (CB) was the best-performing baseline ensemble approach. MEMMS reduced training time by 50%, achieving up to 2 kW lower RMSE, 0.5 kW lower MAE and 5% higher $$R^{2}$$ R 2 for cooling, and up to 2 kW lower RMSE, 4 kW higher MAE and 2% higher $$R^{2}$$ R 2 for heating, showcasing versatility across ETs and forecasting horizons.


  • Top Cited Papers
    Open Access

    Advances in Traffic Congestion Prediction: An Overview of Emerging Techniques and Methods

    by Aristeidis Mystakidis, Paraskevas Koukaras, Christos Tjortjis

    The ongoing increase in urban populations has resulted in the enduring issue of traffic congestion, adversely affecting the quality of life, including commute duration, road safety, and local air quality. Consequently, recognizing and forecasting underlying traffic congestion patterns have become essential, with Traffic Congestion Prediction (TCP) emerging as an increasingly significant area of study. Advancements in Machine Learning (ML) and Artificial Intelligence (AI), as well as improvements in Internet of Things (IoT) sensor technologies have made TCP research crucial to the development of Intelligent Transportation Systems (ITSs). This review examines advanced TCP, emphasizing innovative forecasting methods and technologies and their importance for the ITS sector. This paper provides an overview of statistical, ML, Deep Learning (DL) approaches, and their ensembles that compose TCP. We examine several forecasting methods and discuss relative and absolute evaluation metrics from regression and classification perspectives. Finally, we present an overall step-by-step standard methodology that is often utilized in TCP problems. By combining these elements, this review highlights critical advancements and ongoing challenges in TCP, providing robust and detailed information for state-of-the-art ITS solutions.


  • Open Access

    Optimizing Nurse Rostering: A Case Study Using Integer Programming to Enhance Operational Efficiency and Care Quality

    by Aristeidis Mystakidis, Christos Koukaras, Paraskevas Koukaras, Konstantinos Kaparis, Stavros G. Stavrinides, Christos Tjortjis

    Background/Objectives: This study addresses the complex challenge of Nurse Rostering (NR) in oncology departments, a critical component of healthcare management affecting operational efficiency and patient care quality. Given the intricate dynamics of healthcare settings, particularly in oncology clinics, where patient needs are acute and unpredictable, optimizing nurse schedules is paramount for enhancing care delivery and staff satisfaction. Methods: Employing advanced Integer Programming (IP) techniques, this research develops a comprehensive model to optimise NR. The methodology integrates a variety of constraints, including legal work hours, staff qualifications, and personal preferences, to generate equitable and efficient schedules. Through a case study approach, the model’s implementation is explored within a clinical setting, demonstrating its practical application and adaptability to real-world challenges. Results: The implementation of the IP model in a clinical setting revealed significant improvements in scheduling efficiency and staff satisfaction. The model successfully balanced workload distribution among nurses, accommodated individual preferences to a high degree, and ensured compliance with work-hour regulations, leading to optimised shift schedules that support both staff well-being and patient care standards. Conclusions: The findings underscore the effectiveness of IP in addressing the complexities of NR in oncology clinics. By facilitating a strategic allocation of nursing resources, the proposed model contributes to operational excellence in healthcare settings, underscoring the potential of Operations Research in enhancing healthcare delivery and management practices.


  • Traffic Congestion Prediction: A Machine Learning Approach

    by Olga Geromichalou, Aristeidis Mystakidis, Christos Tjortjis