Search
71 articles for “model robustness”
-
A Comprehensive Investigation of Bagging-Based Ensemble Methods for Improving Machine Learning Model Robustness
Abstract: Machine learning models such as Decision Trees, Logistic Regression, and K-Nearest Neighbors are widely used for classification tasks due to their simplicity and interpretability. However, these models often suffer from high variance, overfitting, and poor generalization when applied to real-world datasets, particularly those that are small, noisy, or imbalanced, as commonly encountered in healthcare, finance, and cybersecurity applications. To address these limitations, this research proposes a Bagging (Bootstrap Aggregating)-based ensemble …
Published in Recent Trends in Mathematics · Vol. 3, Issue 2, 2026 · pp. 24–34 Read article
-
Assessing the Robustness of Machine Learning Models for Wireless Intrusion Detection Under Adversarial Traffic Perturbations
Abstract: As the Internet of Things (IoT) devices and wireless communication networks continue to grow rapidly, protecting systems from cyber threats has become increasingly important. Machine learning–based intrusion detection systems (IDS) have shown strong potential in detecting abnormal and malicious network activities, yet their effectiveness and resilience when facing adversarial attacks are still not sufficiently explored. This research evaluates Machine Learning (ML) models–XGBoost, random forest, and multi-layer perceptron (MLP)—in detecting attacks …
Published in International Journal of Wireless Security and Networks · Vol. 4, Issue 1, 2026 · pp. 29–34 Read article
-
Enhancing Maintenance Decision-Making in Thermal Power Plants Using Generative AI-Based Fault Diagnosis
Abstract: The growing complexity of operation and power consumption of thermal power stations involve the need to have intelligent fault diagnosis systems that can be used to guarantee reliability and safety in operation. In this research, a Generative AI (GenAI)-based hybrid architecture of early fault detection and predictive maintenance is proposed to improve the decision-making process of the maintenance team. The data-driven analytic approach combines methods of data-driven analytics, Generative AI …
Published in International Journal of Energy and Thermal Applications · Vol. 4, Issue 1, 2026 · pp. 25–33 Read article
-
Facial Emotion Detection and Its Applications
Abstract: Facial emotion detection (FED) is an interdisciplinary field that integrates artificial intelligence, computer vision, and machine learning to recognize and interpret human emotions based on facial expressions. The development of FED systems has been propelled by advancements in deep learning, particularly convolutional neural networks (CNNs) and recurrent neural networks (RNNs), which enhance recognition accuracy. Feature extraction techniques, including geometric and appearance-based methods, play a crucial role in classifying emotional states. …
Published in International Journal of Optical Innovations & Research · Vol. 3, Issue 1, 2025 · pp. 8–12 Read article
-
Exploring the Potential of AI-Driven Personalized Learning and Cognitive Interventions for ADHD Management in Indian Children and Adolescents: A Focus on Early Intervention
Abstract: Attention Deficit Hyperactivity Disorder (ADHD) is a prevalent neurodevelopmental condition that significantly impairs attention, executive functioning, and behavioural regulation in children and adolescents. In India, the management of ADHD is particularly challenging due to low public awareness, social stigma, and a critical shortage of specialized mental health services, especially in rural areas. These systemic barriers often lead to delayed diagnoses and limited access to consistent care. However, with the rapid …
Published in International Journal of Brain Sciences · Vol. 2, Issue 2, 2025 · pp. 1–5 Read article
-
QSAR Modeling Techniques: A Comprehensive Review of Tools and Best Practices
Abstract: Quantitative Structure–Activity Relationship (QSAR) modeling has become an essential tool in drug discovery, toxicity assessment, and environmental chemistry. By correlating chemical structure with biological activity or toxicity, QSAR enables the prediction of compound behavior without extensive experimental testing. This approach not only saves time and resources but also supports ethical practices by reducing reliance on animal studies. The evolution of QSAR from basic linear models to advanced machine learning and …
Published in International Journal of Cheminformatics · Vol. 3, Issue 1, 2025 · pp. 56–63 Read article
-
Machine Learning Based Early Cataract Detection: A Predictive Modeling Approach
Abstract: Cataracts, characterized by dense cloudy areas in the eye’s lens, afflict more than 50% of elderly individuals, leading to impaired vision and potential blindness. Detecting cataracts at an early stage is crucial to facilitate simpler treatments, as neglecting the condition may necessitate complex eye surgery. To address this issue, we are creating a predictive system that identifies cataract disease by analyzing user-provided eye features. To achieve this, we leverage OpenCV, …
Published in International Journal of Computer Science Languages · Vol. 1, Issue 2, 2023 · pp. 1–8 Read article
-
The Intersection of Bioinformatics and Cellular Function in Disease Modeling
Abstract: The integration of bioinformatics and cellular biology has revolutionized our understanding of disease mechanisms, offering unprecedented opportunities to model complex biological systems. Bioinformatics is an interdisciplinary field that merges biology, computer science, and statistics, offering advanced tools to analyze vast biological datasets. Cellular functions, including gene expression, protein interactions, and metabolic pathways, form the foundation of physiological and pathological states. Disruptions in these processes can result in diseases like cancer, …
Published in International Journal of Cell Biology and Cellular Functions · Vol. 2, Issue 2, 2024 · pp. 1–7 Read article
-
Integral Sliding Mode Control: A Review of Applications
Abstract: Integral Sliding Mode Control (ISMC) has emerged as a robust and efficient method for handling nonlinear systems with uncertainty, turbulence and external disturbances. This study provides a detailed review of ISMC and its design foundations, design methods, practical application aspects and recent developments are included. ISMC design methodology is explored, to be extended with various design methods with different design methods. Recent advances in research, chatter reduction techniques, applications in …
Published in International Journal of Advanced Control and System Engineering · Vol. 2, Issue 1, 2024 · pp. 25–35 Read article
-
A Systematic Review on Leukemia Detection and Classification Techniques Using Gene Expression
Abstract: Early diagnosis of genetic diseases is crucial for effective treatment, especially in the case of Leukemia, a type of blood cancer characterized by abnormal proliferation of white blood cells. This paper presents a systematic review of recent computational techniques for the detection and classification of Leukemia using gene expression data obtained from DNA microarray analysis. The study explores diverse methodologies including machine learning (ML), deep learning (DL), and bio-inspired algorithms …
Published in International Journal of Genetic Modifications and Recombinations · Vol. 3, Issue 2, 2025 Read article
-
DFT/Data Guided Predictive Modelling of Absorption Maxima in the OLED Rubrene Derivatives
Abstract: This study investigates the optical properties of rubrene derivatives to develop an accurate predictive model for absorption maxima using computational chemistry and chemoinformatic techniques. We benchmarked various quantum chemical methods, identifying that the M06-2X/aug-cc-pVDZ method in dichloromethane (DCM) provided the strongest correlation with experimental data. Key molecular descriptors such as band gap, ionization potential, and electrophilicity index were calculated and analyzed using principal component analysis (PCA) to identify significant factors …
Published in International Journal of Cheminformatics · Vol. 4, Issue 1, 2026 · pp. 41–56 Read article
-
Phase – Field Modeling of Brittle and Ductile Fracture Under Complex Loading Conditions
Abstract: Phase-field modeling has emerged as a powerful computational framework for predicting fracture behavior in engineering materials, offering a unified description of crack initiation, propagation, branching, and coalescence without the need for explicit crack tracking. This study presents an in-depth examination of phase-field modeling applied to both brittle and ductile fracture under complex loading conditions, including multiaxial stress states, cyclic loading, thermal gradients, and dynamic impact. The phase-field approach regularizes the …
Published in International Journal of Fracture Mechanics and Damage Science · Vol. 3, Issue 2, 2025 · pp. 13–18 Read article
-
Remote Sensing and Atmospheric Modelling: Data, Processes, Integration and Future Directions
Abstract: Atmospheric modelling plays a central role in weather forecasting, climate projection, and air quality assessment; however, the availability, accuracy, and representativeness of atmospheric observations fundamentally constrain its reliability. Over the past two decades, rapid advances in remote sensing (RS) have transformed atmospheric observation by providing spatially continuous, multiscale measurements of key atmospheric variables, including aerosols, trace gases, clouds, precipitation, and atmospheric thermodynamic profiles. This review synthesises recent progress in integrating …
Published in International Journal of Atmosphere · Vol. 3, Issue 1, 2026 · pp. 54–67 Read article
-
Bias Detection and Accuracy Enhancement in Voice-based Banking Authentication Using Deep Learning
Abstract: Biometric systems have become an integral part of how many people access banking services today, and voice verification systems can be a secure and easy-to-use source of banking authentication that does not require any physical contact with the bank or any other person. From the security perspective, these systems would normally provide an effective means of identifying an individual but frequently exhibit bias with respect to demographics such as the …
Published in International Journal of Information Security Engineering · Vol. 4, Issue 2, 2026 Read article
-
Fake Product Detection Using Convolutional Neural Networks
Abstract: The widespread circulation of counterfeit products in global markets presents a significant threat to both consumer trust and the integrity of established brands. With the advancement of artificial intelligence, particularly deep learning, there is growing potential to develop more sophisticated systems to combat this issue. This study introduces a novel counterfeit detection framework using the VGG16 Convolutional Neural Network (CNN) to distinguish between authentic and counterfeit products through image analysis. …
Published in International Journal of Algorithms Design and Analysis Review · Vol. 3, Issue 2, 2025 · pp. 08–15 Read article
-
Preserving Saccharomyces cerevisiae Culture for Bee Health and Production
Abstract: Saccharomyces cerevisiae, a single-celled fungus also known as baker’s and brewer’s yeast, is extensively used in brewing, baking, and winemaking due to its robust genetics and utility as a model organism in scientific research. Its role extends beyond these applications into apiculture, where it is utilized to produce supplements and pollen replacements for honeybees. These yeast-based supplements are crucial, especially during times of pollen scarcity or colony stress, as they …
Published in International Journal of Insects · Vol. 1, Issue 2, 2024 · pp. 23–27 Read article
-
QSAR Modeling of a Novel Series of Methoxylated Chalcones as Antioxidant Agents Against Gram-Positive Bacteria Staphylococcus aureus
Abstract: Background: Chalcones are aromatic ketones belonging to the flavonoid family. They are plant-based compounds and are widely found in nature. For centuries, these bioactive molecules have been utilized in various traditional medicines for the treatment of several ailments. Chalcones have antibacterial, antiviral, antimalarial, antifungal, antioxidant, and antileishmanial properties. They are also used to treat inflammation and cancer. Chalcones act as angiogenesis inhibitors, an important factor in cancer progression and metastasis. …
Published in International Journal of Biochemistry and Biomolecule Research · Vol. 3, Issue 1, 2025 · pp. 28–34 Read article
-
Algebraic Foundations of AES (Advanced Encryption Standard): Group Theory and Finite Field Applications in Symmetric Cryptography
Abstract: This paper presents a mathematical study of symmetric cryptographic algorithms, with a particular emphasis on the Advanced Encryption Standard (AES), which is one of the most widely used encryption schemes in modern security applications. The study highlights how abstract mathematical frameworks such as group theory, finite fields, and vector space concepts provide the foundation for the design, implementation, and analysis of AES. By approaching the algorithm from a mathematical perspective, …
Published in Recent Trends in Mathematics · Vol. 2, Issue 1, 2025 · pp. 12–16 Read article
-
Exploring Antimalarial Activity of Chalcone Derivatives through QSAR
Abstract: Background: The core structure of chalcones contains a reactive α,β-unsaturated system within the aromatic rings, which plays a key role in mediating various biological effects. These effects include enzyme inhibition, anticancer activity, anti-inflammatory properties, as well as antibacterial, antifungal, antimalarial, antiprotozoal, and anti-filarial actions.Modifying the structure by introducing substituent groups to the aromatic ring can enhance potency, reduce toxicity, and expand their range of pharmacological actions. Methods: A total of …
Published in International Journal of Bioinformatics and Computational Biology · Vol. 3, Issue 2, 2025 · pp. 1–6 Read article
-
A Detailed Review on Intelligent and Robust Control Strategies for Autonomous Underwater Vehicles with Emphasis on Navigation, Path Tracking, and Stability Enhancement
Abstract: Autonomous Underwater Vehicles (AUVs) have gained significant attention due to their applications in ocean exploration, underwater surveillance, environmental monitoring, and offshore industries. The control of AUVs presents various challenges due to the highly dynamic and uncertain underwater environment, nonlinear hydrodynamics, and external disturbances. This review paper explores various control strategies employed for AUVs, including classical control methods such as Proportional-Integral-Derivative (PID) controllers, modern techniques like Model Predictive Control (MPC), and …
Published in International Journal of Electronics Automation · Vol. 3, Issue 2, 2025 · pp. 28–52 Read article