Traditional Machine Learning Techniques Research Articles

Smart Energy Management in Green Cloud Computing using Machine Learning Algorithms

Cloud computing has many advantages as well as some disadvantages. An internet connection is required to use Cloud Computing. In other words, it is not possible to access the data in cases without internet. Cloud Computing can provide infrastructure services, platform services and software services to individuals with any device connected to the internet. If the connection speed is low when there is internet, the data transmission is also slower. In this context, it may not be practical for individuals or institutions to benefit from Cloud Computing in places where internet connection is low, limited, or absent. A new technology was obtained in this study; this method depends on deep learning and machine learning techniques applied to detect the attacks in the cloud computing-based systems. The suggested method compared with many traditional machine learning techniques.

Automated Cancer Detection Using CNNs: Innovations and Applications

Large-scale datasets and complicated cell architectures provide challenges for traditional tumor identification approaches, which often rely on human analysis or traditional machine-learning techniques. This may result in errors and inefficiencies. The inconsistent shape of tumor cells may be a challenge for these methods, making border detection and classification inconsistent. This paper suggests an automated tumor detection technique based on convolutional neural networks (CNNs) to increase tumor cell border detection and classification accuracy and efficiency. Several experiments were carried out to confirm the suggested strategy's efficacy. This entails developing and refining the CNN architecture in addition to merging deep learning and data augmentation methods. According to the findings, the model has performed well in a variety of experimental settings, especially when it comes to classifying complicated cell samples. Furthermore, the feasibility of using this model for real-time border detection was explored, indicating its suitability for use in a medical setting. Subsequent investigations will concentrate on enhancing the capabilities of models, expanding their range, and tackling problems like disparities in data. This discovery is significant because it has the potential to transform tumor identification by providing a more accurate, scalable, and efficient approach that might be used extensively in clinical settings and eventually lead to better patient outcomes.

Road terrain recognition based on tire noise for autonomous vehicle

Effective road terrain recognition is crucial for enhancing the driving safety, passability, and comfort of autonomous vehicles. This study addresses the challenges of accurately identifying diverse road surfaces using deep learning in complex environments. We introduce a novel end-to-end Tire Noise Recognition Residual Network (TNResNet) integrated with a time-frequency attention module, designed to capture and leverage time-frequency information from tire noise signals for road terrain classification. Our method was evaluated on five distinct road types: asphalt, cement, grass, mud, and sand. The performance of TNResNet was rigorously compared against traditional machine learning techniques, including Decision Trees, K-Nearest Neighbors, and Support Vector Machines, as well as advanced deep learning models like Long Short-Term Memory and Convolutional Neural Networks. Experimental results demonstrate that TNResNet achieves superior classification accuracy of 99.48%, outperforming all comparative methods. This work not only establishes a robust framework for road terrain identification but also showcases the significant practical implications of TNResNet in the realm of autonomous vehicle navigation.

A Comparative Review: Biological Safety and Sustainability of Metal Nanomaterials Without and with Machine Learning Assistance.

In recent years, metal nanomaterials and nanoproducts have been developed intensively, and they are now widely applied across various sectors, including energy, aerospace, agriculture, industry, and biomedicine. However, nanomaterials have been identified as potentially toxic, with the toxicity of metal nanoparticles posing significant risks to both human health and the environment. Therefore, the toxicological risk assessment of metal nanomaterials is essential to identify and mitigate potential adverse effects. This review provides a comprehensive analysis of the safety and sustainability of metallic nanoparticles (such as Au NPs, Ag NPs, etc.) in key domains such as medicine, energy, and environmental protection. Using a dual-perspective analysis approach, it highlights the unique advantages of machine learning in data processing, predictive modeling, and optimization. At the same time, it underscores the importance of traditional methods, particularly their ability to offer greater interpretability and more intuitive results in specific contexts. Finally, a comparative analysis of traditional methods and machine learning techniques for detecting the toxicity of metal nanomaterials is presented, emphasizing the key challenges that need to be addressed in future research.

Survey of SMS Spam Detection Techniques: A Taxonomy

Short Message Service (SMS) spam remains a significant threat to users and businesses, with spammers constantly adopting more sophisticated techniques. This paper comprehensively surveys SMS spam detection methods, categorizing existing approaches into five primary groups: rule-based methods, traditional machine learning techniques, deep learning models, hybrid models, and ensemble methods. Each category is examined in detail, highlighting its strengths, limitations, and evolution. Rule-based methods, though historically significant, are limited by their inability to handle new or evolving spam tactics. Traditional machine learning techniques, such as Naive Bayes and support vector machines (SVM), offer improved accuracy but depend on handcrafted features. In contrast, deep learning models, including recurrent neural networks (RNN) and convolutional neural networks (CNN), excel in feature extraction and adaptability yet face challenges with model complexity and the need for large labeled datasets. Hybrid and ensemble methods combine the benefits of various models to improve performance, reduce bias, and enhance robustness. This review aims to provide a structured overview of the state of SMS spam detection, identify emerging trends, and suggest future research directions, including improving generalization, reducing data dependency, and exploring the integration of contextual information. The findings underscore the need for continued innovation to address the evolving landscape of SMS spam.

Evaluating the Performance of Large Language Models in Predicting Diagnostics for Spanish Clinical Cases in Cardiology

This study explores the potential of large language models (LLMs) in predicting medical diagnoses from Spanish-language clinical case descriptions, offering an alternative to traditional machine learning (ML) and deep learning (DL) techniques. Unlike ML and DL models, which typically rely on extensive domain-specific training and complex data preprocessing, LLMs can process unstructured text data directly without the need for specialized training on medical datasets. This unique characteristic of LLMs allows for faster implementation and eliminates the risks associated with overfitting, which are common in ML and DL models that require tailored training for each new dataset. In this research, we investigate the capacities of several state-of-the-art LLMs in predicting medical diagnoses based on Spanish textual descriptions of clinical cases. We measured the impact of prompt techniques and temperatures on the quality of the diagnosis. Our results indicate that Gemini Pro and Mixtral 8x22b generally performed well across different temperatures and techniques, while Medichat Llama3 showed more variability, particularly with the few-shot prompting technique. Low temperatures and specific prompt techniques, such as zero-shot and Retrieval-Augmented Generation (RAG), tended to yield clearer and more accurate diagnoses. This study highlights the potential of LLMs as a disruptive alternative to traditional ML and DL approaches, offering a more efficient, scalable, and flexible solution for medical diagnostics, particularly in the non-English-speaking population.

Staying ahead of phishers: a review of recent advances and emerging methodologies in phishing detection

The escalating threat of phishing attacks poses significant challenges to cybersecurity, necessitating innovative approaches for detection and mitigation. This paper addresses this need by presenting a comprehensive review of state-of-the-art methodologies for phishing detection, spanning traditional machine learning techniques to cutting-edge deep learning frameworks. The review encompasses a diverse range of methods, including list-based approaches, machine learning algorithms, graph-based analysis, deep learning models, network embedding techniques, and generative adversarial networks (GANs). Each method is meticulously scrutinized, highlighting its rationale, advantages, and empirical results. For instance, deep learning models, particularly convolutional neural networks (CNNs) and recurrent neural networks (RNNs), demonstrate superior detection performance, leveraging their ability to extract complex patterns from phishing data. Ensemble learning techniques and GANs offer additional benefits by enhancing detection accuracy and resilience against adversarial attacks. The impact of this review extends beyond academic discourse, informing practitioners and policymakers about the evolving landscape of phishing detection. By elucidating the strengths and limitations of existing methods, this paper guides the development of more robust and effective cybersecurity solutions. Moreover, the insights gleaned from this review lay the groundwork for future research endeavors, such as integrating contextual information, user behavior analysis, and explainable AI techniques into phishing detection systems. Ultimately, this work contributes to the collective effort to fortify digital defenses against sophisticated phishing threats, safeguarding the integrity of online ecosystems.

Identification of Vertebrae in CT Scans for Improved Clinical Outcomes Using Advanced Image Segmentation

This study proposes a comprehensive framework for the segmentation and identification of vertebrae in CT scans using a combination of deep learning and traditional machine learning techniques. The Res U-Net architecture is employed to achieve a high model accuracy of 93.62% on the VerSe’20 dataset demonstrating effective performance in segmenting lumbar and thoracic vertebrae. Feature extraction is enhanced through the application of Otsu’s method which effectively distinguishes the vertebrae from the surrounding tissue. The proposed method achieves a Dice Similarity Coefficient (DSC) of 87.10% ± 3.72%, showcasing its competitive performance against other segmentation techniques. By accurately extracting vertebral features this framework assists medical professionals in precise preoperative planning, allowing for the identification and marking of critical anatomical features required during spinal fusion procedures. This integrated approach not only addresses the challenges of vertebrae segmentation but also offers a scalable and efficient solution for analyzing large-scale medical imaging datasets with the potential to significantly improve clinical workflows and patient outcomes.

A Comparative Study of Sentiment Analysis on Customer Reviews Using Machine Learning and Deep Learning

Sentiment analysis is a key technique in natural language processing that enables computers to understand human emotions expressed in text. It is widely used in applications such as customer feedback analysis, social media monitoring, and product reviews. However, sentiment analysis of customer reviews presents unique challenges, including the need for large datasets and the difficulty in accurately capturing subtle emotional nuances in text. In this paper, we present a comparative study of sentiment analysis on customer reviews using both deep learning and traditional machine learning techniques. The deep learning models include Convolutional Neural Network (CNN) and Recursive Neural Network (RNN), while the machine learning methods consist of Logistic Regression, Random Forest, and Naive Bayes. Our dataset is composed of Amazon product reviews, where we utilize the star rating as a proxy for the sentiment expressed in each review. Through comprehensive experiments, we assess the performance of each model in terms of accuracy and effectiveness in detecting sentiment. This study provides valuable insights into the strengths and limitations of both deep learning and traditional machine learning approaches for sentiment analysis.

A Time-Frequency-Based Data-Driven Approach for Structural Damage Identification and Its Application to a Cable-Stayed Bridge Specimen.

Structural damage identification based on structural health monitoring (SHM) data and machine learning (ML) is currently a rapidly developing research area in structural engineering. Traditional machine learning techniques rely heavily on feature extraction, where weak feature extraction can lead to suboptimal features and poor classification performance. In contrast, ML-based methods, particularly deep learning approaches like convolutional neural networks (CNNs), automatically extract relevant features from raw data, improving the accuracy and adaptability of the damage identification process. This study developed a time-frequency-based data-driven approach aiming to improve the effectiveness of traditional data-driven structural damage identification approaches for large complex structures. Firstly, the structural acceleration signals in the time domain were converted into two-dimensional images via the Gram angle difference field (GADF). Subsequently, the characteristic feature in the image data was studied by convolutional neural networks (CNNs) to predict the structural damage conditions. An experimental study on a scale model of a cable-stayed bridge was conducted to identify the damage of stay cables under the moving vehicle load on the main girders. The CNN was employed to extract the characteristic features from the time-varying monitoring data of vehicle-bridge interactions. The CNN parameters were optimized to conduct the structural damage classification task. The performance of the proposed method was evaluated by comparing it with various traditional pre-trained networks. The effect of environmental noise on the prediction accuracy was also investigated. The numerical results show that the ResNet model has the best performance in terms of damage identification accuracy and convergence speed, achieving higher accuracy and faster convergence compared to the other four traditional networks. The method can accurately identify damage on bridges using insufficient sensors on the bridge deck, which has valuable potential for application to real-world bridges with monitoring data. As the Signal-to-Noise Ratio (SNR) decreases from 20 dB to 2.5 dB, the prediction accuracy of ResNet decreases from 86.63% to 62.5%, which demonstrates the robustness and reliability in identifying structural damage.

Surrounding Rock Squeezing Classification in Underground Engineering Using a Hybrid Paradigm of Generative Artificial Intelligence and Deep Ensemble Learning

Surrounding rock squeezing is a common geological disaster in underground excavation projects (e.g., TBM tunneling and deep mining), which has adverse effects on construction safety, schedule, and property. To predict the squeezing of the surrounding rock accurately and quickly, this study proposes a hybrid machine learning paradigm that integrates generative artificial intelligence and deep ensemble learning. Specifically, conditional tabular generative adversarial network is devised to solve the problems of data shortage and class imbalance for data augmentation at the data level, and the deep random forest is built based on the augmented data for subsequent squeezing classification. A total of 139 historical squeezing cases are collected worldwide to validate the efficacy of the proposed modeling paradigm. The results reveal that this paradigm achieves a prediction accuracy of 92.86% and a macro F1-score of 0.9292. In particular, the individual F1-scores on strong squeezing and extremely strong squeezing are more than 0.9, with excellent prediction reliability for high-intensity squeezing. Finally, a comparative analysis with traditional machine learning techniques is conducted and the superiority of this paradigm is further verified. This study provides a valuable reference for surrounding rock squeezing classification under a limited data environment.

Application of Convolutional Neural Networks for Advanced Classification of Power Quality Signals

In this work, we explore the classification of power quality signals using a simulated database incorporating both unique and combined disturbances in accordance with IEEE 1159 standards. The primary focus is on employing convolutional neural networks (CNNs) for the detection and classification of these anomalies. To assess the effectiveness of our approach, we conducted a comprehensive comparison between the outcomes from CNNs and traditional machine learning techniques, as well as with recent literature findings. Preliminary results indicate that CNNs exhibit a remarkable ability to capture distinctive features of power quality signals, surpassing traditional methodologies in terms of accuracy and robustness. This study not only reaffirms the potential of convolutional neural networks in the field of power quality monitoring but also paves the way for future investigations that might explore more complex deep learning approaches to further enhance the accuracy of anomaly classification in electrical systems.

An optimized ensemble model with advanced feature selection for network intrusion detection.

In today's digital era, advancements in technology have led to unparalleled levels of connectivity, but have also brought forth a new wave of cyber threats. Network Intrusion Detection Systems (NIDS) are crucial for ensuring the security and integrity of networked systems by identifying and mitigating unauthorized access and malicious activities. Traditional machine learning techniques have been extensively employed for this purpose due to their high accuracy and low false alarm rates. However, these methods often fall short in detecting sophisticated and evolving threats, particularly those involving subtle variations or mutations of known attack patterns. To address this challenge, our study presents the "Optimized Random Forest (Opt-Forest)," an innovative ensemble model that combines decision forest approaches with genetic algorithms (GAs) for enhanced intrusion detection. The genetic algorithms based decision forest construction offers notable benefits by traversing a wider exploration space and mitigating the risk of becoming stuck in local optima, resulting in the discovery of more accurate and compact decision trees. Leveraging advanced feature selection techniques, including Best-First Search, Particle Swarm Optimization (PSO), Evolutionary Search, and Genetic Search (GS), along with contemporary dataset, this research aims to enhance the adaptability and resilience of NIDS against modern cyber threats. We conducted a comprehensive evaluation of the proposed approach against several well-known machine learning models, including AdaBoostM1 (AbM1), K-nearest neighbor (KNN), J48-Decision Tree (J48), multilayer perceptron (MLP), stochastic gradient descent (SGD), naïve Bayes (NB), and logistic model tree (LMT). The comparative analysis demonstrates the effectiveness and superiority of our method across various performance metrics, highlighting its potential to significantly enhance the capabilities of network intrusion detection systems.

Intelligent open-set MIMO recognition in OWC using a Siamese neural network.

Multiple-input multiple-output (MIMO) technology, a core component of 6G, has been widely adopted in optical wireless communication (OWC) systems. Accurate recognition of different MIMO types is essential for MIMO selection and demodulation. In this Letter, we propose an open-set MIMO recognition method for OWC systems using a Siamese neural network (SNN). Simulation results show that the SNN significantly outperforms other recognition approaches, including convolutional neural networks (CNNs) and traditional machine learning techniques. For SNN-based recognition, over 90% accuracy is achieved with training based on only nine fixed sampling points in both 2 × 2 and 4 × 4 MIMO-OWC systems.

An Efficient Flow-Based Anomaly Detection System for Enhanced Security in IoT Networks.

The growing integration of Internet of Things (IoT) devices into various sectors like healthcare, transportation, and agriculture has dramatically increased their presence in everyday life. However, this rapid expansion has exposed new vulnerabilities within computer networks, creating security challenges. These IoT devices, often limited by their hardware constraints, lack advanced security features, making them easy targets for attackers and compromising overall network integrity. To counteract these security issues, Behavioral-based Intrusion Detection Systems (IDS) have been proposed as a potential solution for safeguarding IoT networks. While Behavioral-based IDS have demonstrated their ability to detect threats effectively, they encounter practical challenges due to their reliance on pre-labeled data and the heavy computational power they require, limiting their practical deployment. This research introduces the IoT-FIDS (Flow-based Intrusion Detection System for IoT), a lightweight and efficient anomaly detection framework tailored for IoT environments. Instead of employing traditional machine learning techniques, the IoT-FIDS focuses on identifying unusual behaviors by examining flow-based representations that capture standard device communication patterns, services used, and packet header details. By analyzing only benign traffic, this network-based IDS offers a streamlined and practical approach to securing IoT networks. Our experimental results reveal that the IoT-FIDS can accurately detect most abnormal traffic patterns with minimal false positives, making it a feasible security solution for real-world IoT implementations.

Detection of advanced persistent threat: A genetic programming approach

Advanced Persistent Threats (APTs) are an intimidating class of cyberattacks known for their persistence, sophistication, and targeted nature. These attacks, coordinated by highly motivated adversaries, pose a grave risk to organizations and individuals, often operating stealthily and evading detection. While existing research primarily focuses on applying Machine Learning (ML) methods to analyze network traffic data for APT detection, this article introduces a novel approach that utilizes Genetic Programming (GP). The proposed method not only detects APT attacks but also identifies their specific life cycle stages through the evolutionary capabilities of GP. Its effectiveness lies in its ability to excel in detecting intricate patterns, even within classes with a limited number of instances, a feat that is often challenging for traditional ML techniques. The method involves evolving and optimizing its models to effectively learn and adapt to complex APT behaviors. Experimentation with a publicly available dataset showcases the efficacy of the proposed method across diverse APT stages. The results demonstrate that the proposed method, GPC, achieves a 3.71% improvement in balanced accuracy compared to the best-performing model from related works. Moreover, a thorough analysis of the best-evolved GP model uncovers valuable insights about identified features and significant patterns. This research advances the APT detection paradigm by leveraging GP’s capabilities, providing a fresh and effective perspective on countering these persistent threats.

Enhancing water security through advanced modeling: integrating deep learning and a novel metaheuristic optimization algorithm for accurate pan evaporation prediction

ABSTRACT This study evaluates and enhances machine learning models for predicting pan evaporation under diverse climatic conditions. Five fundamental machine learning models were employed and tested across four different stations. Subsequent comparisons were made with advanced techniques, including long short-term memory (LSTM) networks. An innovative approach was introduced, combining LSTM with Binary Al-Biruni Earth Radius (BER–LSTM). This hybrid method was benchmarked against other optimization techniques. The BER–LSTM model consistently outperformed other models across all stations and time scales, achieving up to a 97.54% improvement in root mean square error (RMSE) compared to standard LSTM on daily time scales. Compared to simpler models like Multilayer Perceptron and Support Vector Regressor, BER–LSTM showed even more substantial improvements, with up to a 99.03% reduction in RMSE. The BER–LSTM model demonstrates superior predictive capabilities for pan evaporation across varied climatic conditions, offering significant improvements over both traditional and advanced machine learning techniques. This approach shows promise for enhancing evaporation forecasting in diverse environmental contexts.

Prediction of Temperature Distribution on an Aircraft Hot-Air Anti-Icing Surface by ROM and Neural Networks

To address the inefficiencies and time-consuming nature of traditional hot-air anti-icing system designs, reduced-order models (ROMs) and machine learning techniques are introduced to predict anti-icing surface temperature distributions. Two models, AlexNet combined with Proper Orthogonal Decomposition (POD-AlexNet) and multi-CNNs with GRU (MCG), are proposed by comparing several classic neural networks. Design variables of the hot-air anti-icing cavity are used as inputs of the two models, and the corresponding surface temperature distribution data serve as outputs, and then the performance of these models is evaluated on the test set. The POD-AlexNet model achieves a mean prediction accuracy of over 95%, while the MCG model reaches 96.97%. Furthermore, the proposed model demonstrates a prediction time of no more than 5.5 ms for individual temperature samples. The proposed models not only provide faster predictions of anti-icing surface temperature distributions than traditional numerical simulation methods but also ensure acceptable accuracy, which supports the design of aircraft hot-air anti-icing systems based on optimization methods such as genetic algorithms.

LSTM Algorithm in Predicting Chronic Kidney Disease Optimized Using Genetic Algorithm

Chronic Kidney Disease is a health condition in which the kidneys experience a progressive decline in function. Kidneys are vital organs that filter waste and excess blood fluids. CKD can lead to excess products in the body and cause various health issues, so early detection of CKD is necessary. While traditional machine learning techniques have performed well in predicting CKD in existing studies, this study investigates the potential of long short-term memory (LSTM) optimized with Genetic Algorithm to enhance predictive accuracy and efficiency by optimizing its hyperparameters, including number of units, hidden layers, activation function, recurrent activation, and dropout rate. The result demonstrates that the optimized LSTM slightly performs better than without optimization, achieving higher precision, recall, accuracy, and f1 score by 100% respectively. This outstanding result can be attributed to several key factors, such as ensuring rigorous data preprocessing and utilizing k-fold cross-validation to make the model more reliable. This indicates the hybrid approach can be a powerful method for the early detection of CKD, leading to better patient outcomes. Despite the promising performance, further research is suggested, specifically using a larger dataset to ensure applicability to more general population and exploring other optimization methods to reduce computational cost.

Edge computing-based ensemble learning model for health care decision systems

A growing number of humans have suffered severe chronic illnesses, which has caused a boost in the requirement for diagnostic and medical treatment procedures that are both accurate and fast. Improved patient conditions and enhanced Decision-Making Systems (DMS) for healthcare professionals are the primary objectives of the Clinical Decision Support System (CDSS) recommended in this research article. The main drawback of traditional Machine Learning (ML) techniques is their failure to predict reliably. To solve this problem, the proposed model creates an Ensemble Extreme Learning Machine (EN-ELM) algorithm that combines predictors trained on several different data sets. This lowers the chance of overfitting. The suggested CDSS uses many different data processing methods, including Adaptive Synthetic (ADASYN) and isolation Forest (iForest), which fix problems like outliers and class imbalance. This approach significantly enhances the framework’s classification performance. Also, the CDSS is compatible with an EC model, which enables real-time computation while minimizing the requirement for integrated systems. The recommended CDSS applies iForest and ADASYN to execute large-scale trials validating high standards of accuracy across numerous datasets. Researchers concluded that a suitable ELM classification threshold of 85% is the most effective, which substantially boosts the accuracy of the predictive model. When applied to various medical datasets, such as Hepatocellular Carcinoma (HCC), Cervical Cancer, Chronic Kidney Disease (CKD), Heart Disease, and Arrhythmia, the EN-ELM achieved accuracy rates of 99.36%, 98.15%, 97.85%, 97.06%, and 96.72%, respectively. By measuring this progress, the CDSS could dramatically improve the accuracy of chronic illness diagnosis and treatment, which similarly affects clinicians.