Advanced electroencephalogram based authentication by adapting machine learning models

The surgeons of the future face serious challenges. A major concern is the reduction in surgical training time. For example, the European Working Time Directive has reduced surgical training time by two thirds. The strong response from senior surgeons in the United States to the reduction in surgical training time to just over 80 hours a week (even though this is still nearly twice the European quota) reflects the gravity of the situation. In the United Kingdom, this reduction in training time is exacerbated by the extra foundation year after graduation, which has resulted in an additional year …

Purpose–This paper presents a comprehensive empirical analysis focusing on sentiment flux within state-of-the-art models designed for handling polarity shifts due to implicit negation in Amazon mobile phones' reviews. Method–The research evaluates diverse models across five categories: traditional machine learning (ML), deep learning (DL), and hybrid models combining both approaches. Various feature extraction, feature selection, and data augmentation techniques are tested on Amazon mobile phone reviews dataset. BERT and LSTM are used for deep learning while SVM and Naive Bayes are used for traditional ML. ANOVA is used to identify the presence or absence of significant differences and interactions among these entities. Results –DL shows superior performance compared to traditional ML models. ANOVA analysis shows significant performance differences between conventional ML and DL models. Traditional ML models interact significantly with feature extraction and selection techniques while DL models do not. Traditional ML models do not interact significantly with data augmentation methods while DL models do. FastText extraction outperforms word2vec; Back translation outperforms synonym replacement while recursive feature selection (RFE) surpasses TF-IDF (Term Frequency-Inverse Document Frequency). The BERT and LSTM exhibit one of the strongest performances. Conclusion –The study concludes that DL models are more effective. Data augmentation techniques significantly impact the performance of DL models, with back translation showing superior performance over synonym replacement. This provides a leverage point in developing an improved model in the future. Recommendations –Future research should focus on developing a hybrid model for Enhanced Polarity Shift Management of Mobile Phone Reviews using Contextual Back Translation Augmented by Seq2seq Perturbations. This aims at leveraging contextual back translation and Seq2seq perturbations to generate a diverse interpretation that consequently improves the model's ability to handle nuanced expressions of sentiments due to implicit negation with enhanced accuracy, generalizability, robustness to polarity shifts, and contextual understanding. Research Implications –The findings provide valuable insights into the development of state-of-the-art models, offering a promising direction for further research in sentiment analysis. Keywords –empirical analysis, hybrid, perturbations, implicit negation, sentiment flux

: Asynchronous training technologies enable students to master material much more rapidly than classroom instruction. Students' time saving represents a major potential benefit of using them. This paper fills a methodology gap in estimating the students' timesaving benefit of asynchronous training technologies. Meta- analyses of their effectiveness yield a statistic called Effect Size. Estimating the benefit in dollar terms requires information regarding the reduction in total training time. This paper presents a methodology for inferring the percentage reduction in students' training time from estimates of Effect Size. It goes on to infer the percentage reductions for three asynchronous training technologies from estimates of their Effect Sizes. Finally, it compares the estimates to some direct measures of training time savings. Asynchronous training technologies enable students to master material much more rapidly than classroom instruction. Students' time saving represents a major potential benefit of using them. This paper fills a methodology gap in estimating the students' timesaving benefit of asynchronous training technologies. Meta- analyses of their effectiveness yield a statistic called Effect Size. Estimating the benefit in dollar terms requires information regarding the reduction in total training time. This paper presents a methodology for inferring the percentage reduction in students' training time from estimates of Effect Size. It goes on to infer the percentage reductions for three asynchronous training technologies from estimates of their Effect Sizes. Finally, it compares the estimates to some direct measures of training time savings.

Recognition of objects from EEG signals requires selection of appropriate feature extraction and classification techniques with best efficiency in terms of highest classification accuracy with lowest run time for its applications in real time. The objective of this paper is to analyze the performance of various feature extraction techniques and to choose that particular method which can be implemented in real time system with best efficiency. The EEG signals are acquired from subjects while they explored the objects visually and visuo-tactually. Thus acquired EEG signals are preprocessed followed by feature extraction using adaptive autoregressive (AAR) parameters, ensemble empirical mode decomposition (EEMD), approximate entropy (ApEn) and multi-fractal detrended fluctuation analysis (MFDFA). The performance of these features are analyzed in terms of their dimension, extraction time and also depending upon the classification results produced by three classifiers [Support Vector machine (SVM), Naive Bayesian (NB), and Adaboost (Ada)] independently according to classification accuracy, sensitivity and classification times. The experimental results show that AAR parameter has an optimum dimension of 36 (not too large like EEMD i.e. 7,680 or too small like ApEn i.e. 6) and required minimum extraction as well as classification time of 0.59 and 0.008 s respectively. AAR also yielded highest maximum classification accuracy and sensitivity of 80.95 and 92.31 % respectively with NB classifier. Thus AAR parameters can be chosen for real time object recognition from EEG signal along with Naive Bayesian classifier.

This paper analyzes the optimization features of machine learning (ML) model training procedures using multi-GPU systems to enhance cyber security in telecommunication networks. A key aspect of the study is the use of data parallelism, which allows the distribution of the training load across multiple GPUs, significantly reducing training time and improving model accuracy-critical factors for rapid threat detection in cyberspace. A novel approach for optimizing data batch size using Mutual Information (MI) is proposed, which harmonizes the utilization of computational resources with the information content of the training data. MI helps to determine the optimal data batch size that minimizes training errors and improves model accuracy without a significant increase in training time. Experimental results demonstrate the substantial advantages of multi-GPU configurations compared to single-GPU setups, providing faster training and improved model accuracy. It was particularly emphasized that MI-guided batch size tuning significantly outperforms traditional manual tuning methods, ensuring higher validation accuracy and reducing training time. The study showed that the MI-based approach is an effective tool for addressing the problem of optimizing ML model training processes in real-world scenarios where cyber security is critical. The proposed methods allow ML models to train faster and more accurately identify potential threats, making them particularly relevant for telecommunication networks where a rapid response to new threats in real time is required. The implementation of modern computational technologies such as multi-GPU systems and MI-optimized training enhances the efficiency and accuracy of machine learning models. This, in turn, improves cyber security measures and ensures a more reliable defence of telecommunication networks against malicious attacks. It is noted that the proposed approaches can be adapted not only for cyber security but also for other areas where high model accuracy and fast training are important. Future research prospects include the development of new machine learning methods, particularly deep neural networks, the exploration of alternative computational architectures such as quantum computing or distributed systems, and their integration into real-time systems. Special attention should be paid to the ethical aspects of implementing automated cyber security systems, particularly in preventing bias in algorithms and ensuring fairness in their application.

Individualized drug dosing is a highly effective strategy for optimizing therapeutic outcomes, especially for drugs with high inter-individual variability. Population pharmacokinetic modeling is a widely used approach to characterize inter-individual variability in therapeutic drug monitoring. However, the development of population pharmacokinetic models is labor intensive and requires significant technical expertise. Machine learning (ML) represents a promising alternative for personalized drug dosing strategies. Despite numerous studies applying ML in this context, no previous work has comprehensively reviewed and compared their methodologies and predictive performance. This scoping review addresses this gap in the existing literature with the aim to examine the methodological approaches used in ML-based pharmacokinetic modeling for dose optimization. Five databases were systematically searched from their inception to May 2025. Studies comparing predictions of drug concentrations or pharmacokinetic parameters between ML and population pharmacokinetic models were included. Studies published in non-English language, reviews, protocols, or studies that did not employ ML models for individualized dose regimens or treatment plans were excluded. Fifty-eight studies were included. We found that boosting-based models, tree-based models, instance-based, and regression-based models were the most commonly used ML approaches. Approximately 31% of the studies integrated ML with population pharmacokinetic models, while the remainder developed stand-alone ML models. Inconsistencies in reporting were evident, as only 60% of the studies detailed their feature selection methods. Model evaluation approaches also varied: 47% of ML models used internal test sets, while the remainder employed external datasets or mixed approaches. In terms of predictive accuracy, ML models performed comparably to or better than population pharmacokinetic models, especially for drugs with significant pharmacokinetic variability. This review identifies substantial heterogeneity in ML modeling approaches, feature selection, and model evaluation. To enhance the reproducibility and clinical applicability of ML models in individualized drug dosing, standardization in reporting and methodological practices is essential.

Adversarial examples have become one of the largest challenges that machine learning models, especially neural network classifiers, face. These adversarial examples break the assumption of attack-free scenario and fool state-of-the-art (SOTA) classifiers with insignificant perturbations to human. So far, researchers achieved great progress in utilizing adversarial training as a defense. However, the overwhelming computational cost degrades its applicability and little has been done to overcome this issue. Single-Step adversarial training methods have been proposed as computationally viable solutions, however they still fail to defend against iterative adversarial examples. In this work, we first experimentally analyze several different SOTA defense methods against adversarial examples. Then, based on observations from experiments, we propose a novel single-step adversarial training method which can defend against both single-step and iterative adversarial examples. Lastly, through extensive evaluations, we demonstrate that our proposed method outperforms the SOTA single-step and iterative adversarial training defense. Compared with ATDA (single-step method) on CIFAR10 dataset, our proposed method achieves 35.67% enhancement in test accuracy and 19.14% reduction in training time. When compared with methods that use BIM or Madry examples (iterative methods) on CIFAR10 dataset, it saves up to 76.03% in training time with less than 3.78% degeneration in test accuracy.

Machine learning models, particularly ensemble methods such as bagging, often require substantial computational resources and long training times, especially with large datasets. This work presents a parallel bagging approach that leverages shared memory parallelism on multi core systems to reduce training time while maintaining high resource efficiency. Classifier profiling is used to estimate computational costs, and a linear programming based allocation optimizer distributes CPU cores proportionally to these estimates. Experiments on the Covertype dataset demonstrate significant reductions in training time, notable speedup, and balanced core utilization across different hardware configurations. The results highlight the importance of optimizing both software design and hardware usage to fully exploit parallelism.

Internet of Things (IoT) devices are widely used but also vulnerable to cyberattacks that can cause security issues. To protect against this, machine learning approaches have been developed for network intrusion detection in IoT. These often use feature reduction techniques like feature selection or extraction before feeding data to models. This helps make detection efficient for real-time needs. This paper thoroughly compares feature extraction and selection for IoT network intrusion detection in machine learning-based attack classification framework. It looks at performance metrics like accuracy, f1-score, and runtime, etc. on the heterogenous IoT dataset named Network TON-IoT using binary and multiclass classification. Overall, feature extraction gives better detection performance than feature selection as the number of features is small. Moreover, extraction shows less feature reduction compared with that of selection, and is less sensitive to changes in the number of features. However, feature selection achieves less model training and inference time compared with its counterpart. Also, more space to improve the accuracy for selection than extraction when the number of features changes. This holds for both binary and multiclass classification. The study provides guidelines for selecting appropriate intrusion detection methods for particular scenarios. Before, the TON-IoT heterogeneous IoT dataset comparison and recommendations were overlooked. Overall, the research presents a thorough comparison of feature reduction techniques for machine learning-driven intrusion detection in IoT networks.

Diabetic Retinopathy (DR) is a health condition caused due to Diabetes Mellitus (DM). It causes vision problems and blindness due to disfigurement of human retina. According to statistics, 80% of diabetes patients battling from long diabetic period of 15 to 20 years, suffer from DR. Hence, it has become a dangerous threat to the health and life of people. To overcome DR, manual diagnosis of the disease is feasible but overwhelming and cumbersome at the same time and hence requires a revolutionary method. Thus, such a health condition necessitates primary recognition and diagnosis to prevent DR from developing into severe stages and prevent blindness. Innumerable Machine Learning (ML) models are proposed by researchers across the globe, to achieve this purpose. Various feature extraction techniques are proposed for extraction of DR features for early detection. However, traditional ML models have shown either meagre generalization throughout feature extraction and classification for deploying smaller datasets or consumes more of training time causing inefficiency in prediction while using larger datasets. Hence Deep Learning (DL), a new domain of ML, is introduced. DL models can handle a smaller dataset with help of efficient data processing techniques. However, they generally incorporate larger datasets for their deep architectures to enhance performance in feature extraction and image classification. This paper gives a detailed review on DR, its features, causes, ML models, state-of-the-art DL models, challenges, comparisons and future directions, for early detection of DR.

Due to the vital role of the aberrant DNA methylation during the disease development such as cancer, the comprehension of its mechanism had become essential in the recent years for early detection and diagnosis. With the advent of the high-throughput technologies, there are still several challenges to achieve the classification process using the DNA methylation data. The high-dimensionality and high-noisiness of the DNA methylation data may lead to the degradation of the prediction accuracy. Thus, it becomes increasingly important in a wide range to employ robust computational tools such as feature selection and extraction methods to extract the informative features amongst thousands of them, and hence improving cancer prediction. By using the DNA methylation degree in promoters and probes regions, this paper aims at predicting cancer with a hybridized approach based on the feature selection and feature extraction techniques. The suggested approach exploits a filter feature selection method called ( F-score ) to overcome the high-dimensionality problem of the DNA methylation data, and proposes an extraction model which employs the peaks of the mean methylation density, the fast Fourier transform algorithm, and the symmetry between the methylation density of a sample and the mean methylation density of both sample types normal and cancer as novel feature extraction methods, in order to accurate cancer classification and reduce training time. To evaluate the reliability of our approach, The naive base, random forest, and support vector machine algorithms are introduced to predict different cancer types like: breast, colon, head, kidney, lung, thyroid, and uterine with and without the hybridized approach. The results show that, the classification accuracy improves in all most cases and it also proves the reliability indirectly.

Accurate classification of skin diseases remains a significant challenge due to the wide range of conditions, image variability, and class imbalance. Conventional diagnostic systems often struggle with limited accuracy and computational inefficiency, hindering real-time clinical use. This study presents a hybrid deep learning model that integrates ResNet-50 with an Improved Whale Optimization Algorithm (IWOA) to address these issues. A diverse skin image dataset was collected and preprocessed before being passed through ResNet-50 for feature extraction. IWOA was employed to optimize key parameters, enhancing model training and convergence. Experimental results show that the proposed IWOA-ResNet model achieves 99.09% accuracy, with a 25% reduction in training time, maintaining strong performance across unbalanced and varied data. When compared to traditional CNN and machine learning models, the hybrid approach demonstrates superior accuracy and efficiency. This research highlights the potential of combining deep learning with metaheuristic optimization for automated, real-time skin disease diagnosis, offering a scalable and robust solution for clinical deployment.

We live in the era of large data analysis, where processing vast datasets has become essential for uncovering valuable insights across various domains of our lives. Machine learning (ML) algorithms offer powerful tools for processing and analyzing this abundance of information. However, the considerable time and computational resources needed for training ML models pose significant challenges, especially within cascade schemes, due to the iterative nature of training algorithms, the complexity of feature extraction and transformation processes, and the large sizes of the datasets involved. This paper proposes a modification to the existing ML-based cascade scheme for analyzing large biomedical datasets by incorporating principal component analysis (PCA) at each level of the cascade. We selected the number of principal components to replace the initial inputs so that it ensured 95% variance retention. Furthermore, we enhanced the training and application algorithms and demonstrated the effectiveness of the modified cascade scheme through comparative analysis, which showcased a significant reduction in training time while improving the generalization properties of the method and the accuracy of the large data analysis. The improved enhanced generalization properties of the scheme stemmed from the reduction in nonsignificant independent attributes in the dataset, which further enhanced its performance in intelligent large data analysis.

A novel ensemble local graph structure based feature extraction network for EEG signal analysis

Feature extraction for machine learning-based intrusion detection in IoT networks