Conventional Machine Research Articles

WITHDRAWN: CSCAD: An adaptive LightGBM algorithm to detect cache side-channel attacks

Cache side-channel attacks have become more sophisticated and more destructive to the security of computer architectures than ever before, which can obtain the plaintext of a victim without key. To address the challenges introduced by cache side-channel attacks, anomaly detection and feature detection were proposed. However, these methods have drawbacks in terms of computational performance and detection effectiveness. In this paper, we proposed a Cache Side-Channel Attack Detector (CSCAD), a novel tool for detecting cache side-channel attacks against memory events. Specifically, we design a collector using Hardware Performance Counters and use improved Maximum Information Coefficient to generate feature vectors. Meanwhile, an adaptive genetic algorithm with crossover and mutation probability is proposed to optimize the hyperparameters of LightGBM. Additionally, an adaptive loss function weight model with low overhead is introduced to enhance the efficiency of attack detection. It is encouraging to see that CSCAD achieved a recall of 98.14%. In detecting 1000 samples, it boosted the detection speed by approximately 75% compared to conventional machine learning methods. CSCAD has outperformed the state-of-the-art methods by simultaneously achieving excellent detection speed and effectiveness.

How Interpretable Machine Learning Can Benefit Process Understanding in the Geosciences

AbstractInterpretable Machine Learning (IML) has rapidly advanced in recent years, offering new opportunities to improve our understanding of the complex Earth system. IML goes beyond conventional machine learning by not only making predictions but also seeking to elucidate the reasoning behind those predictions. The combination of predictive power and enhanced transparency makes IML a promising approach for uncovering relationships in data that may be overlooked by traditional analysis. Despite its potential, the broader implications for the field have yet to be fully appreciated. Meanwhile, the rapid proliferation of IML, still in its early stages, has been accompanied by instances of careless application. In response to these challenges, this paper focuses on how IML can effectively and appropriately aid geoscientists in advancing process understanding—areas that are often underexplored in more technical discussions of IML. Specifically, we identify pragmatic application scenarios for IML in typical geoscientific studies, such as quantifying relationships in specific contexts, generating hypotheses about potential mechanisms, and evaluating process‐based models. Moreover, we present a general and practical workflow for using IML to address specific research questions. In particular, we identify several critical and common pitfalls in the use of IML that can lead to misleading conclusions, and propose corresponding good practices. Our goal is to facilitate a broader, yet more careful and thoughtful integration of IML into Earth science research, positioning it as a valuable data science tool capable of enhancing our current understanding of the Earth system.

Fusion of multi-temporal PlanetScope data and very high-resolution aerial imagery for urban tree species mapping

Detailed assessment of the ecosystem services provided by urban green spaces requires data on urban tree species. While many approaches for mapping of urban trees from remotely sensed data have been proposed, the fusion of multi-temporal satellite imagery with very high resolution orthophotos remains relatively underexplored. In this research, we assess the potential of a multimodal deep learning approach with intermediate data fusion for classifying common tree species found in the Brussels Capital Region. Our method combines two image sources: (a) multi-temporal PlanetScope data and (b) high-resolution aerial imagery. To evaluate the contribution of each image source, we separately train and assess each branch of the network. Both image sources demonstrate the ability to predict prevalent tree species with high accuracy. However, the fusion of the two image sources yields the best results, achieving an overall accuracy of 0.88 for the five most common tree species in the region. Our approach is compared to two conventional machine learning methods: a random forest (RF) and a support vector machine classifier (SVM) and outperforms both with a 11 percentage point increase in overall accuracy over RF and a 10 percentage point increase over SVM. Increasing the number of species from five to thirteen, including all species with more than 500 tree samples, results in a marginal decrease in accuracy (from 0.88 to 0.84). Overall, our deep learning approach demonstrates its efficacy in classifying common tree species in urban settings and provides a foundation for a comprehensive quantification of ecosystem services offered by urban trees through remote sensing data.

Physics-informed surrogate modeling for a damaged rotating shaft

In the online structural health monitoring framework, surrogate modeling aims to approximate the predictions of the underlying high-fidelity model to facilitate fast simulations. Generally, conventional machine learning techniques based on purely data-driven approaches often behave as black boxes, producing predictions that may lack physical consistency. In an effort to overcome these limitations, we proposed a neural network mapping constrained by the known physical relationships of an engineering model, which can be considered as a physics-informed inductive bias. This approach aims to enhance computational efficiency while reducing the required number of training examples. The physics-informed term guides the training process to yield predictions that exhibit greater physical consistency while maintaining high performance. We here propose a surrogate modeling approach relying on a Physics-Informed Neural Network (PINN) scheme, which is designed to enable robust and real-time monitoring of ballistic impact damage on a rotor shaft. At first, a high fidelity finite element model is developed to simulate the response of the rotor shaft under various operational conditions. As a first reduction step, substructuring is used to decrease the dimensionality of the system's matrices, thereby reducing the computational cost. For comparison and validation, an alternative model based on the Timoshenko beam theory is utilized. Both models examine the effects of impact-induced vibration loads due to ballistic occurrences. The PINN model receives the external load as input and predicts the system's response. The system matrices extracted from both models constitute the physics-informed term. This term is incorporated into the PINN architecture in order to steer the learning process towards more physically consistent and reliable predictions regarding the state and the response of the structure. The algorithm's performance is evaluated using response data derived from both models.

Analysis of the relevance and prospects of application of federate training

This article examines federated learning (FOE) as an innovative approach to machine learning, different from traditional methods. In conventional machine learning (MO), data is collected on a central server to train the model. However, in the case of FO, the learning model is directed to data distributed across local devices, and learning takes place directly on these devices. In addition, the article discusses methods and algorithms of federated learning, identifies the advantages and real areas of application of federated learning. FO is used in various fields, including working with medical data and personal data of customers in sales companies. This approach is especially valuable for ensuring data confidentiality and privacy.

Blockchain Abstract Architecture for Energy Collaboration for Agriculture in Indonesia based on Bibliometric Approach

The direction of development for the application of electric vehicles is starting to be seen with great certainty, this situation is driven by the vision of sustainable development goals (SDGs) where one of them is energy that is minimal from carbon. The agricultural sector is a field of activity that also requires machinery or equipment to support its activities. and currently most of them still use conventional machines that produce quite high carbon, meanwhile, when agricultural machinery is converted to electricity, the challenge is to provide affordable energy at competitive prices and reliable technology. Blockchain is a disruptive technology that has many advantages, especially in terms of flexibility and security to accommodate broad interactions from many parties in an activity chain. The purpose of this research is to develop a blockchain-based architecture to support collaborative energy supply in agriculture that is sourced from renewable energy. we use Scopus indexed data sources to formulate and describe the directions and references of the proposed architecture

A Novel Hybrid Feature Extraction Methods for EEG Based Emotions Classification Using Ensemble Approaches in HIED Dataset

Introduction: Emotions play a significant role in human relationships in everyday challenges. Human-machine interactions and the complexities of its research include humanoid robots which rely significantly on emotional queries. The present study recommends a novel methodology for human emotion identification using electroencephalogram (EEG) signals. Machine learning (ML) techniques based on electroencephalogram have demonstrated impressive results in distinguishing emotions. Method: The present study investigates five ensemble learning-based machine learning (EML) and five conventional machine learning (CML) algorithms for achieving the recognition of human emotions from EEG signals. The main objective of the current research study is to implement the optimal approach for classifying emotions and subsequently will achieve it by using physiological characteristics to categorise the seven main emotional states (stress, happy, sadness, anger, fear, disgust, and surprise). This research follows the study investigation of ensemble learning-based machine learning (EML) and conventional machine learning (CML) algorithms for achieving the recognition of human emotions from EEG signals. Initially, the EEG data in this study are categorized into theta, alpha, beta, and gamma bands using a DWT. Result: With this idea, the research elaborates on further decomposition based on bandseparated EEG signals known as intrinsic mode functions (IMFs) along with empirical mode decomposition is subsequently implemented. Then, 13 statistical characteristics are extracted from the IMFs using five multiclass EML algorithms: adaptive boosting, rotation random forest, bagging, random forest, and extreme gradient boost. The following procedure is carried out to construct an ML-based system. In the final phase, 10-fold cross-validation is used to assess the efficacy of these five EML algorithms. Accuracy, F1-score, and area-under-the-curve (AUC) are performance assessment metrics that are employed to compare the final results of these algorithms with the five CML techniques. Conclusion: The proposed method for EEG-based emotion analysis is significantly more effective and achieves greater outcomes than all of the techniques that were compared, based on the data from the experiments.

MediaPipe with GNN for Human Activity Recognition

Human interaction and computer vision converge in the realm of Human Activity Recognition (HAR), which is a research field dedicated to the creation of automated systems capable of observing and categorizing human activities. This domain closely aligns with machine learning, involving the development of algorithms and models adept at learning to recognize and classify patterns within data. HAR typically unfolds in two pivotal phases: data acquisition and processing, followed by activity classification. In the initial phase of data acquisition and processing, information is gathered from various sensors or video sources, such as accelerometers, smartphones, or smartwatches. Subsequently, the collected data undergo preprocessing to extract relevant features. The subsequent phase, activity classification, employs machine learning algorithms to categorize these extracted features into distinct activity types, ranging from walking and running to sitting. This paper introduces an innovative approach grounded on these two fundamental phases. For the first phase, we leverage the MediaPipe algorithm to discern human articulations. Once these poses are detected, we contribute by extracting the coordinates of each articulation. These coordinates are then transformed into graphs, where nodes signify the articulation coordinates and edges represent the connections between them. In the second phase, we enhance existing methodologies by incorporating a diverse set of machine learning models. Notably, the utilization of Graph Neural Networks (GNNs) which stands out as a significant advancement. This choice proves instrumental in effectively learning and representing complex spatial and temporal patterns, surpassing the limi-tations of conventional machine learning algorithms. The developed system undergoes evaluation on the KTH and UCF50 datasets, demonstrating state-of-the-art performance in HAR.

Deep learning-based estimation of time-dependent parameters in Markov models with application to nonlinear regression and SDEs

We present a novel deep-learning method for estimating time-dependent parameters in Markov processes through discrete sampling. Departing from conventional machine learning, our approach reframes parameter approximation as an optimization problem using the maximum likelihood approach. Experimental validation focuses on parameter estimation in multivariate regression and stochastic differential equations (SDEs). Theoretical results show that the real solution is close to SDE with parameters approximated using our neural network derived under specific conditions. Our work contributes to SDE-based model parameter estimation, offering a versatile tool for diverse fields.

LLM is All You Need : How Do LLMs Perform On Prediction and Classification Using Historical Data

This study investigates the utility of large language models (LLMs) in performing traditional machine learning tasks such as prediction, and explores the potential of refinement architectures to enhance their effectiveness in these roles. Utilizing the Titanic survival dataset, we conducted a com- parative analysis using both conventional machine learning tools and LLM-based approaches. Our findings indicate that while LLMs differ fundamentally from traditional ML models in prediction tasks, there exist specific architectural modifications, termed Thought Refinement Architectures, which can significantly improve their performance. These results highlight the potential for inte- grating LLMs into traditional ML workflows, thereby expanding their applicability and enhancing predictive accuracy.

Meta-2OM: A multi-classifier meta-model for the accurate prediction of RNA 2'-O-methylation sites in human RNA.

2'-O-methylation (2-OM or Nm) is a widespread RNA modification observed in various RNA types like tRNA, mRNA, rRNA, miRNA, piRNA, and snRNA, which plays a crucial role in several biological functional mechanisms and innate immunity. To comprehend its modification mechanisms and potential epigenetic regulation, it is necessary to accurately identify 2-OM sites. However, biological experiments can be tedious, time-consuming, and expensive. Furthermore, currently available computational methods face challenges due to inadequate datasets and limited classification capabilities. To address these challenges, we proposed Meta-2OM, a cutting-edge predictor that can accurately identify 2-OM sites in human RNA. In brief, we applied a meta-learning approach that considered eight conventional machine learning algorithms, including tree-based classifiers and decision boundary-based classifiers, and eighteen different feature encoding algorithms that cover physicochemical, compositional, position-specific and natural language processing information. The predicted probabilities of 2-OM sites from the baseline models are then combined and trained using logistic regression to generate the final prediction. Consequently, Meta-2OM achieved excellent performance in both 5-fold cross-validation training and independent testing, outperforming all existing state-of-the-art methods. Specifically, on the independent test set, Meta-2OM achieved an overall accuracy of 0.870, sensitivity of 0.836, specificity of 0.904, and Matthew's correlation coefficient of 0.743. To facilitate its use, a user-friendly web server and standalone program have been developed and freely available at http://kurata35.bio.kyutech.ac.jp/Meta-2OM and https://github.com/kuratahiroyuki/Meta-2OM.

Machine translation of Chinese classical poetry: a comparison among ChatGPT, Google Translate, and DeepL Translator

Recent studies have highlighted ChatGPT’s remarkable capabilities in machine translation. However, little attention has been paid to its application in literary translation, particularly within the realm of Chinese classical poetry. To explore the potential of ChatGPT’s abilities in poetry translation, we conducted a comparative analysis of poetry translation quality, contrasting ChatGPT (with two different prompts) with Google Translate and DeepL Translator regarding fidelity, fluency, language style, and machine translation style. The results revealed that ChatGPT outperformed Google Translate and DeepL Translator in all evaluation criteria, suggesting its exceptional ability in poetry translation. Furthermore, when employing a prompt that instructs ChatGPT to preserve the rhythm and rhyme of poems, ChatGPT demonstrated a remarkable ability to retain the beauty of the original poetic language, setting itself apart from conventional machine translation systems. Our analysis further elucidated ChatGPT’s proficiency in comprehending and translating some common symbols, imagery, and underlined semantic components which contributes to coherent and fluent translations. Our research opens up ChatGPT’s new possibilities in translating ancient literary texts into foreign languages.

Using deep learning model to predict micronutrients in food after processing

Understanding the nutritional content of food after processing is important for the food processing industry. Choosing the appropriate processing method allows users to retain healthy micronutrients. In fact, collecting nutrient information of food before and after processing poses many challenges due to biological changes and interactions of nutritional components. Currently, the approach is to collect data on each nutritional component before and after processing. Conventional machine learning models will then use this data to produce good prediction results but with limited stability. Therefore, we proposed to use a deep learning model to conduct training on a data set with 27 nutritional components that change through two processing processes, boil and fry, extracted from a standard reference data set of the United States. The results show accurate prediction and improve the stability by 8.6%. The study shows the potential of improving deep learning models in predicting post-processing nutritional composition in food processing.

Attenuated total reflectance-flourier transformed infrared spectroscopy (ATR-FTIR) coupled with deep learning: A rapid method for geographical origin identification of sea cucumber Apostichopus japonicus

The sea cucumber Apostichopus japonicus is one of the China’s most prized seafoods, its geographical origin is a decisive factor for its economic value. To quickly and effectively identify the geographical origin of this species, ATR-FTIR spectroscopy coupled with deep learning methods was applied in the present study. Compared to conventional machine learning models (support vector machine, random forest and lightGBM), the proposed one-dimensional convolutional neural network (1D-CNN) model demonstrates significant advantages in terms of data automation and accuracy. The model can correctly classify sea cucumbers from different sea regions up to 90.7%, among which 100% identification of Apostichopus japonicus from the East China Sea can be realized. The results proved that ATR-FTIR spectroscopy combined with 1D-CNN could be used as a rapid and effective technique for tracing the geographical origin of Apostichopus japonicus.

Enhancing trustworthiness among iot network nodes with ensemble deep learning-based cyber attack detection

A lot of machine learning methods and expert systems are used in network intrusion detection automation. When different industrial control systems merge with the Internet of Things (IoT) environment, they become vulnerable to cyber-attacks in critical infrastructure situations requiring communication technologies. Conventional machine learning techniques used for network anomaly detection are ineffective due to the substantial amount of network traffic within important Cyber-Physical Systems (CPSs). In this manuscript, Cyberattack Identification Through Ensemble Deep Learning in an IoT environment is proposed. Initially, the input network traffic data are taken from the IoT-23 dataset. Then the network traffic data are preprocessed using Z-Score normalization to reduce any irrelevant or erroneous data from the input dataset. Then the relevant features are selected using the Gorilla Troops Optimization (GTO) algorithm. Afterwards, the selected features are fed into the ensemble classification model based on Random Space (RS), Random Tree (RT), Extreme Gradient Boosting (XGBoost), and Graph Convolutional Neural Network (GCNN). Among the several ensembling techniques, GCNN can achieve better performance. Python is used to accomplish the suggested technique. The performance of the proposed GCNN method provides 12.09%, 4.34%, and 3.21% higher accuracy than the other models like RS, RT and XGBoost respectively.

Enhancing Malware Detection by Integrating Machine Learning with Cuckoo Sandbox

In this work, two categories of deep learning and conventional machine learning were used to classify malware using a dataset of all possible API call sequences. Specifically, the objective was to determine the best strategy to tackle the ever-rising menace as malware becomes more complex. A new dataset was created employing Cuckoo Sandbox, where API call sequences originating from both benign and malware samples were recorded. The performance of these algorithms was benchmarked and tested using this dataset, which includes SVM, RF, KNN, XGB, GBC, CNN, and RNN. The study established that both deep learning and conventional machine learning algorithms provided high accuracy above 90%. Specifically, the recurrent neural networks (RNNs) demonstrated high accuracy rates ranging from 95% to 99%. These results are highly indicative of deep learning, especially RNN, as a promising approach to improving the effectiveness of malware detection. The data obtained from dynamic analysis, when integrated into a database, serves as a more reliable source for training and testing of such models, and can improve the model’s ability to identify new threats posed by malware. Thus, this work is salient in enhancing the development of new approaches to fight malware that constantly evolve in the modern world.

Improving skin cancer detection by Raman spectroscopy using convolutional neural networks and data augmentation.

Our previous studies have demonstrated that Raman spectroscopy could be used for skin cancer detection with good sensitivity and specificity. The objective of this study is to determine if skin cancer detection can be further improved by combining deep neural networks and Raman spectroscopy. Raman spectra of 731 skin lesions were included in this study, containing 340 cancerous and precancerous lesions (melanoma, basal cell carcinoma, squamous cell carcinoma and actinic keratosis) and 391 benign lesions (melanocytic nevus and seborrheic keratosis). One-dimensional convolutional neural networks (1D-CNN) were developed for Raman spectral classification. The stratified samples were divided randomly into training (70%), validation (10%) and test set (20%), and were repeated 56 times using parallel computing. Different data augmentation strategies were implemented for the training dataset, including added random noise, spectral shift, spectral combination and artificially synthesized Raman spectra using one-dimensional generative adversarial networks (1D-GAN). The area under the receiver operating characteristic curve (ROC AUC) was used as a measure of the diagnostic performance. Conventional machine learning approaches, including partial least squares for discriminant analysis (PLS-DA), principal component and linear discriminant analysis (PC-LDA), support vector machine (SVM), and logistic regression (LR) were evaluated for comparison with the same data splitting scheme as the 1D-CNN. The ROC AUC of the test dataset based on the original training spectra were 0.886±0.022 (1D-CNN), 0.870±0.028 (PLS-DA), 0.875±0.033 (PC-LDA), 0.864±0.027 (SVM), and 0.525±0.045 (LR), which were improved to 0.909±0.021 (1D-CNN), 0.899±0.022 (PLS-DA), 0.895±0.022 (PC-LDA), 0.901±0.020 (SVM), and 0.897±0.021 (LR) respectively after augmentation of the training dataset (p<0.0001, Wilcoxon test). Paired analyses of 1D-CNN with conventional machine learning approaches showed that 1D-CNN had a 1-3% improvement (p<0.001, Wilcoxon test). Data augmentation not only improved the performance of both deep neural networks and conventional machine learning techniques by 2-4%, but also improved the performance of the models on spectra with higher noise or spectral shifting. Convolutional neural networks slightly outperformed conventional machine learning approaches for skin cancer detection by Raman spectroscopy.

Image-based malware analysis for enhanced IoT security in smart cities

To develop safer smart city solutions, it is crucial to investigate new technologies for malware analysis and detection to enhance existing malware prevention systems. Deep learning (DL) techniques have surpassed conventional machine learning as the dominant method for network security; therefore, it is crucial for researchers to utilise DL techniques to address the velocity, volume, and complexity of modern malware, as these approaches can effectively handle large amounts of data and extract representative information robustly. This study investigates the application of deep learning techniques, specifically different autoencoders (AEs), to improve malware analysis and prevention in Internet of Things (IoT)-based smart cities. This research evaluates different convolutional neural network based AE structures and aims to establish a robust malware analysis method by focusing on image classification. By experimenting on both greyscale and RGB malware imagery datasets, it has been demonstrated that variational AE can not only detect nearly all malware but also illustrate the generalisability and efficacy of AE in addressing malware threats. The study systematically evaluates different AE configurations, and this comparative analysis can inspire further research into deep learning techniques for IoT security measures.

Hand Gesture Recognition Based on High-Density Myoelectricity in Forearm Flexors in Humans.

Electromyography-based gesture recognition has become a challenging problem in the decoding of fine hand movements. Recent research has focused on improving the accuracy of gesture recognition by increasing the complexity of network models. However, training a complex model necessitates a significant amount of data, thereby escalating both user burden and computational costs. Moreover, owing to the considerable variability of surface electromyography (sEMG) signals across different users, conventional machine learning approaches reliant on a single feature fail to meet the demand for precise gesture recognition tailored to individual users. Therefore, to solve the problems of large computational cost and poor cross-user pattern recognition performance, we propose a feature selection method that combines mutual information, principal component analysis and the Pearson correlation coefficient (MPP). This method can filter out the optimal subset of features that match a specific user while combining with an SVM classifier to accurately and efficiently recognize the user's gesture movements. To validate the effectiveness of the above method, we designed an experiment including five gesture actions. The experimental results show that compared to the classification accuracy obtained using a single feature, we achieved an improvement of about 5% with the optimally selected feature as the input to any of the classifiers. This study provides an effective guarantee for user-specific fine hand movement decoding based on sEMG signals.

IMG-11. RADIOGENOMIC ANALYSIS INFORMS ON IMMUNOLOGICAL PROFILES, PROGNOSIS, AND TREATMENT RESPONSE IN PEDIATRIC LOW-GRADE GLIOMA

Abstract BACKGROUND Pediatric low-grade gliomas (pLGGs) have variable prognosis and treatment responses. Complete resection cannot be achieved for all tumors, especially for highly infiltrative or deep-seated tumors, necessitating additional therapy, from chemotherapy to targeted inhibitors. We integrated imaging-derived phenotypes with genotypic traits from transcriptional analysis, offering an in-depth characterization of pLGG immune microenvironment, progression risk, and likelihood of multiple treatments. METHODS Analyzing 549 treatment-naïve pLGGs with multiparametric MRI and RNA sequencing, we identified distinct immunological groups using XCell scores based on immune cell infiltration. We developed a radiomic signature using conventional MRI and machine learning techniques (support vector machines with a linear kernel and nested cross-validation) to distinguish the ‘immune-hot’ group, and incorporated diffusion MRI to improve signature accuracy. Additionally, a clinicoradiomic model predicting tumor progression risk and treatment response was trained, integrating clinical and radiomic data. Transcriptomic analysis was conducted to identify pathways correlated with clinicoradiomic risk, predictive of pLGG progression. RESULTS Three immunological groups were revealed, the ‘immune-hot’ group characterized by poor prognosis due to a high concentration of pro-tumorigenic M2-polarized macrophages, despite a higher preponderance of T-lymphocytes. The radiomic signature effectively distinguished the ‘immune-hot’ group with balanced accuracies of 76.8%/86.0% in discovery/replication sets, improved by diffusion MRI to 81.5%/84.4%. The clinicoradiomic model showed concordance indices of 0.71 (discovery) and 0.77 (replication), predicting patient progression risk. Significant differences (p=0.0010) were found in clinicoradiomic risk scores between patients undergoing one versus multiple treatments post-diagnosis, linking higher scores to a likelihood of multiple treatments. Transcriptomic pathways associated with higher clinicoradiomic risk highlighted the importance of fatty acid oxidation, a tumor-promoting mechanism that drives adaptive resistance to cytolytic immune cell effectors. CONCLUSIONS This first large-scale radiogenomic analysis in pLGGs aids in prognostication, assessing progression risk, predicting treatment response to standard-of-care therapies, and stratification of patients to identify potential candidates for novel therapies targeting aberrantly regulated pathways.