Machine Learning-based System Research Articles

Embedded machine learning-based road conditions and driving behavior monitoring

Car accident rates have increased in recent years, resulting in losses in human lives, properties, and other financial costs. An embedded machine learning-based system is developed to address this critical issue. The system can monitor road conditions, detect driving patterns, and identify aggressive driving behaviors. The system is based on neural networks trained on a comprehensive dataset of driving events, driving styles, and road conditions. The system effectively detects potential risks and helps mitigate the frequency and impact of accidents. The primary goal is to ensure the safety of drivers and vehicles. Collecting data involved gathering information on three key road events: normal street and normal drive, speed bumps, circular yellow speed bumps, and three aggressive driving actions: sudden start, sudden stop, and sudden entry. The gathered data is processed and analyzed using a machine learning system designed for limited power and memory devices. The developed system resulted in 91.9% accuracy, 93.6% precision, and 92% recall. The achieved inference time on an Arduino Nano 33 BLE Sense with a 32-bit CPU running at 64 MHz is 34 ms and requires 2.6 kB peak RAM and 139.9 kB program flash memory, making it suitable for resource-constrained embedded systems.

Detection of Phishing Websites by Using URL Analysis

Abstract: In recent years, with the increasing use of mobile devices, there is a growing trend to move almost all real-world operations to the cyberworld. Although this makes easy our daily lives, it also brings many security breaches due to the anonymous structure of the Internet. Used antivirus programs and firewall systems can prevent most of the attacks. However, experienced attackers target on the weakness of the computer users by trying to phish them with bogus webpages. These pages imitate some popular banking, social media, e-commerce, etc. sites to steal some sensitive information such as, user-ids, passwords, bank account, credit card numbers, etc. Phishing detection is a challenging problem, and many different solutions are proposed in the market as a blacklist, rule-based detection, anomaly-based detection, etc. In the literature, it is seen that current works tend on the use of machine learning-based anomaly detection due to its dynamic structure, especially for catching the “zero-day” attacks. In this paper, we proposed a machine learning-based phishing detection system by using eight different algorithms to analyse the URLs, and three different datasets to compare the results with other works. The experimental results depict that the proposed models have an outstanding performance with a success rate.

Enhancing resilience and security in the U.S. power grid against cyber-physical attacks

The U.S. power grid faces increasing risks from cyber-physical attacks that could disrupt essential services and compromise national security. The fusion of artificial intelligence with cybersecurity protocols is perceived to present a consequential shift in the current efforts to safeguard critical infrastructures. This manuscript aims to identify the most critical vulnerabilities within the power grid's infrastructure, develop advanced machine learning-based threat detection systems, and propose automated response mechanisms to mitigate impacts effectively. By integrating comprehensive vulnerability assessments, innovative detection technologies, and autonomous response strategies, this study seeks to enhance the resilience of the power grid against sophisticated cyber threats.

Development of Random Forest Model for Stroke Prediction

Stroke is a significant cause of mortality and morbidity worldwide, and early detection and prevention of stroke are essential for improving patient outcomes. Machine learning algorithms have been used in recent years to predict the risk of stroke by leveraging large amounts of clinical and demographic data. The development of a stroke prediction system using Random Forest machine learning algorithm is the main objective of this thesis. The primary goal of the project is to increase the accuracy of stroke detection while addressing the shortcomings of the current system, which include real- time deployment and interpretability issues with logistic regression. The development and use of an ensemble machine learning-based stroke prediction system, performance optimization through the use of ensemble machine learning algorithms, performance assessment, and real-time model deployment through the use of Python Django are among the goals of the research. The study's potential to improve public health by lessening the severity and consequences of strokes through early diagnosis and treatment makes it significant. Data collection, preprocessing, model selection, evaluation, and real-time deployment using Python Django are all part of the research technique. Our dataset consists of 5110 rows of tuples and columns with total size of 69kg. The performance of our stroke prediction algorithm was evaluated using confusion metrics-consisting of accuracy, precision, recall and F1-score. At the end of the research, Random Forest model gave an accuracy of 98.5% compared to the existing model logistic regression which has 86% accuracy.

Data-Driven ICS Network Simulation for Synthetic Data Generation

Data-Driven ICS Network Simulation for Synthetic Data Generation

Wind speed estimation and maximum power point tracking using neuro-fuzzy systems for variable-speed wind generator

This paper proposes a novel method using a machine learning-based Adaptive Neuro-Fuzzy Inference System (ANFIS) to optimize Maximum Power Point Tracking (MPPT) in variable-speed Wind Turbines (WT). The ANFIS algorithm, blending artificial neural networks and fuzzy logic, addresses issues with traditional wind speed sensors, such as cost, imprecision, and susceptibility to adverse weather conditions. An initial offline-trained ANFIS is suggested to understand turbine power characteristics, and subsequently estimate varying wind speed, addressing strong nonlinearity due to WT aerodynamics and wind speed fluctuations. A second ANFIS efficiently tracks the maximum power point, overcoming limitations of linear controllers. Implemented in Matlab/Simulink for a 3.5 kW WT, the approach demonstrates effectiveness, precision, and faster response time in wind speed estimation and accurate MPPT compared to alternatives. A notable advantage is its independence from instantaneous wind speed measurement, providing a cost-effective solution for wind energy systems.

The Design of Machine Learning-Based Computer-Aided System with LabVIEW For Abnormalities in Mammogram Images

Mammogram is the best way of breast cancer detection nowadays, as breast cancer is the most common form of cancer in the female gender and this form of cancer usually causes death. Many scientists, doctors, and engineers are working together to deal with such serious issues in human life. This paper, it is aimed to develop a new computer-aided system with a graphical coded language to detect abnormalities in mammogram images by using machine learning technics such as ANN and SVM. The developed algorithm has a graphical user interface (GUI) and all results are shown in there. The algorithm was created using three different stages. These are image processing and mass segmentation, feature selection and extraction, and classification. To test the accuracy of the system as the sensitivity, specificity, and accuracy, mammogram images with forty benign and forty malignant masses were used. The obtained results for measuring the sensitivity, specificity, and accuracy are 95%, 97.5%, and 96.25% for ANN and 97.5%, 97.5, and 97.5 for SVM, respectively. As can be said that the algorithm, user-friendly due to its user interface, can be preferred because it can detect many cancerous cells such as breast cancer with high accuracy.

ROSPaCe: Intrusion Detection Dataset for a ROS2-Based Cyber-Physical System and IoT Networks

Most of the intrusion detection datasets to research machine learning-based intrusion detection systems (IDSs) are devoted to cyber-only systems, and they typically collect data from one architectural layer. Often the attacks are generated in dedicated attack sessions, without reproducing the realistic alternation and overlap of normal and attack actions. We present a dataset for intrusion detection by performing penetration testing on an embedded cyber-physical system built over Robot Operating System 2 (ROS2). Features are monitored from three architectural layers: the Linux operating system, the network, and the ROS2 services. The dataset is structured as a time series and describes the expected behavior of the system and its response to ROS2-specific attacks: it repeatedly alternates periods of attack-free operation with periods when a specific attack is being performed. This allows measuring the time to detect an attacker and the number of malicious activities performed before detection. Also, it allows training an intrusion detector to minimize both, by taking advantage of the numerous alternating periods of normal and attack operations.

Forecasting cryptocurrency's buy signal with a bagged tree learning approach to enhance purchase decisions.

The cryptocurrency market is captivating the attention of both retail and institutional investors. While this highly volatile market offers investors substantial profit opportunities, it also entails risks due to its sensitivity to speculative news and the erratic behavior of major investors, both of which can provoke unexpected price fluctuations. In this study, we contend that extreme and sudden price changes and atypical patterns might compromise the performance of technical signals utilized as the basis for feature extraction in a machine learning-based trading system by either augmenting or diminishing the model's generalization capability. To address this issue, this research uses a bagged tree (BT) model to forecast the buy signal for the cryptocurrency market. To achieve this, traders must acquire knowledge about the cryptocurrency market and modify their strategies accordingly. To make an informed decision, we depended on the most prevalently utilized oscillators, namely, the buy signal in the cryptocurrency market, comprising the Relative Strength Index (RSI), Bollinger Bands (BB), and the Moving Average Convergence/Divergence (MACD) indicator. Also, the research evaluates how accurately a model can predict the performance of different cryptocurrencies such as Bitcoin (BTC), Ethereum (ETH), Cardano (ADA), and Binance Coin (BNB). Furthermore, the efficacy of the most popular machine learning model in precisely forecasting outcomes within the cryptocurrency market is examined. Notably, predicting buy signal values using a BT model provides promising results.

Performance of a machine learning-based early warning system in hospitalized patients with diabetes mellitus

Performance of a machine learning-based early warning system in hospitalized patients with diabetes mellitus

Developing an IoT and Machine Learning-Based Monitoring System for Discrete Production Processes

Developing an IoT and Machine Learning-Based Monitoring System for Discrete Production Processes

ARBUR, a machine learning-based analysis system for relating behaviors and ultrasonic vocalizations of rats

ARBUR, a machine learning-based analysis system for relating behaviors and ultrasonic vocalizations of rats

Smart Trashbin using Machine learning and Internet of Things

Health and hygiene are vital areas in the ever-expanding community that cannot be disregarded or compromised for the welfare of human communities. This idea uses the Internet of Things to separate biodegradable and non-biodegradable waste into smart dustbins. People can no longer separate their waste into biodegradable and non-biodegradable categories because of their hectic work schedules. By automatically identifying waste as biodegradable or non biodegradable, smart dustbins with image recognition technology improve waste management by encouraging effective waste segregation and lessening the load on landfills. The goal of an IOT and machine learning-based waste management system for a green, smart society is to use the newest IOT technologies to improve the efficiency of trash management from every apartment in the community.

New Tiger Beetle Algorithm for Cybersecurity, Medical Image Segmentation and Other Global Problems Optimization

The tiger beetle is a fierce and cunning predator insect that uses deception to hunt its prey. The tiger beetle traps and hunts them by digging holes along the path of other insects. This study has used the tiger beetle's hunting strategy to create the tiger beetle optimization (TBO) algorithm. In this algorithm, each solution represents the position of a tiger beetle, with the optimal position being the prey's location. Using this method, the tiger beetles gradually converge to the optimal solution, creating holes around them and searching for them. We evaluate the TBO algorithm's search capability using a series of well-known mathematical test functions. Moreover, Among the sophisticated forms of malware are polymorphic viruses, which are adept at changing their behaviour while maintaining the same essential functions. Thus, a machine learning-based malware analysis system utilizing the power of the proposed TBO is introduced in this article. Compared to other optimization methods, the proposed algorithm has shown less error in finding the optimal solution when implemented and evaluated on different functions. The tiger beetle optimization algorithm has proven helpful in various applications, including image clustering and reservoir well placement, where it can identify damaged areas or tissues with greater accuracy. When diagnosing lung cancer, the proposed method has shown a sensitivity, validity, and accuracy of 88.63\%, 87.58\%, and 89.86\%, respectively, using EBT, WKNN, ESKNM, and QSVM methods.

Developing MLP based prediction system for anticancer drug response using hybrid features of genomics and cheminformatics

Traditional cancer treatment methods have become less effective due to the increasing diversity of cancer types. To address this, precision medicine has gained support within the medical community. This approach tailors treatment to individual patients based on their specific disease characteristics. However, a major challenge lies in accurately predicting how a patient will respond to a specialized drug. Numerous machine learning-based predictive systems have been developed to address this challenge. These systems utilize genomic signatures and the chemical structure of drugs to predict drug activity. In this paper, we introduce a Multi-Layer Perceptron (MLP) based system for predicting the response of anticancer drugs. Our system utilizes hybrid features derived from both genetic expression and the chemical structure of drugs. It is developed using the well-known GDSC dataset (Genomics of Drug Sensitivity in Cancer). Our system achieved a lower Root Mean Square Error (RMSE) value of 0.889, in contrast to the RMSE value of 0.983 obtained by the current state-of-the-art (SOTA) system, SwNet. This indicates superior predictive accuracy. The findings suggest that our proposed research holds promise for the development of targeted drugs for anticancer treatments.

An Intelligent Computer Aided Diagnosis System for Classification of Ovarian Masses using Machine Learning Approach

Ovarian cancer, a difficult and often asymptomatic malignancy, remains a substantial global health concern in women. An ovary is a female reproductive organ, which lies on each side of the uterus and used to store eggs. Computer-aided diagnosis (CAD) is an approach that involves using computer algorithms and machine learning techniques to assist medical professionals in diagnosing ovarian malignancies, benign tumors or Poly-cystic ovaries (PCOS). The need for models that can effectively predict benign ovarian tumors and ovarian cancer has led to the use of machine learning techniques. Our research objective is to propose a machine learning-based system for accurate and early ovarian mass detection utilizing novel annotated ovarian masses. We have used an actual patient database whose input features were extracted from 187 transvaginal ultrasound images from database. The input image is preprocessed using the Block Matching 3D filter. The process involves employing binary and watershed segmentation techniques, followed by the integration of Gabor, Gray-Level Co-Occurrence Matrix (GLCM), Tamura, and edge feature extraction methods. K-Nearest Neighbors (KNN) and Random Forest (RF) are two classifiers used for classification. Based on our results, we are able to demonstrate that binary segmentation with RF classifiers is more accurate (above 86%) than KNN classifiers (under 84%).

Creation of Machine Learning-Based Fish Classification Systems Based on Morphometric and Mathematical Transform Data

The accurate classification of fish species is crucial for various ecological, conservation, and fisheries management applications. Traditional methods relying solely on manual observation or morphological characteristics can be time-consuming and prone to errors. In recent years, advancements in machine learning (ML) techniques coupled with morphometric and mathematical transform data have shown promising results in automating and enhancing fish species classification processes. This paper presents a comprehensive review of the development and application of ML-based fish classification systems utilizing morphometric measurements and mathematical transforms. The core of this paper focuses on the integration of morphometric and mathematical transform data with ML algorithms for fish classification. In this review commonly employed ML algorithms, including but not limited to support vector machines (SVM), artificial neural networks (ANN), random forests (RF), and convolutional neural networks (CNN). Furthermore, this work discuss feature selection techniques to optimize classification performance and reduce dimensionality. These may include the incorporation of multimodal data sources, transfer learning approaches, and the development of user-friendly tools for field biologists and conservationists. Overall, this paper serves as a comprehensive guide for researchers and practitioners interested in leveraging ML techniques for accurate and efficient fish species classification.

Tunable Q-factor wavelet transform based identification of diabetic patients using ECG signals

Diabetes is a chronic health condition that is characterized by increased levels of glucose (sugar) in the blood. It can have harmful effects on different parts of the body, such as the retina of the eyes, skin, nervous system, kidneys, and heart. Diabetes affects the structure of electrocardiogram (ECG) impulses by causing cardiovascular autonomic dysfunction. Multi-resolution analysis of the input ECG signal is utilized in this paper to develop a machine learning-based system for the automated detection of diabetic patients. In the first step, the input ECG signal is decomposed into sub-bands utilizing the tunable Q-factor wavelet transform (TQWT) technique. In the second step, four entropy-based characteristics are evaluated from each SB and elected using the K-W test method. To develop an automatic diabetes detection system, selected features are given as input with 10-fold validation to a SVM classifier using various kernel functions. The 3 rd sub-band of TQWT with the Coarse Gaussian kernel function kernel of the SVM classifier yields a classification accuracy of 91.5%. In the same dataset, the comparative analysis demonstrates that the proposed method outperforms other existing methods.

Identification of Depression from the Speech of a Person

The world’s population has experienced profound psychological effects over time due to stress, anxiety, and fast-paced modern lifestyles. Accurate mapping of the diverse forms of human biology is made possible by the global technological development in healthcare, which digitizes scopious data more than usual measuring methods. Healthcare data analysis can now be done more effectively with the use of machine learning (ML), thanks to its vast volume of data. To forecast the likelihood of mental illnesses and, consequently, carry out possible treatment outcomes, machine learning techniques are being applied to the field of mental health. This review paper outlines many machine learning techniques that are employed in the identification of depression. Classification, Deep learning, and ensemble—are used to group the machine learning-based depression detection systems. A comprehensive approach for identification depression is introduced, which includes data extraction, pre-processing, ML classifier training, detection, classification, and performance assessment. Additionally, it provides a summary that identifies the goals and constraints of various research projects that have been presented in the field of depression detection. It also covers potential directions for future research in the realm of depression diagnosis

Empirical Enhancement of Intrusion Detection Systems: A Comprehensive Approach with Genetic Algorithm-based Hyperparameter Tuning and Hybrid Feature Selection

AbstractMachine learning-based IDSs have demonstrated promising outcomes in identifying and mitigating security threats within IoT networks. However, the efficacy of such systems is contingent on various hyperparameters, necessitating optimization to elevate their performance. This paper introduces a comprehensive empirical and quantitative exploration aimed at enhancing intrusion detection systems (IDSs). The study capitalizes on a genetic algorithm-based hyperparameter tuning mechanism and a pioneering hybrid feature selection approach to systematically investigate incremental performance improvements in IDS. Specifically, our work proposes a machine learning-based IDS approach tailored for detecting attacks in IoT environments. To achieve this, we introduce a hybrid feature selection method designed to identify the most salient features for the task. Additionally, we employed the genetic algorithm (GA) to fine-tune hyperparameters of multiple machine learning models, ensuring their accuracy in detecting attacks. We commence by evaluating the default hyperparameters of these models on the CICIDS2017 dataset, followed by rigorous testing of the same algorithms post-optimization through GA. Through a series of experiments, we scrutinize the impact of combining feature selection methods with hyperparameter tuning approaches. The outcomes unequivocally demonstrate the potential of hyperparameter optimization in enhancing the accuracy and efficiency of machine learning-based IDS systems for IoT networks. The empirical nature of our research method provides a meticulous analysis of the efficacy of the proposed techniques through systematic experimentation and quantitative evaluation. Consolidated in a unified manner, the results underscore the step-by-step enhancement of IDS performance, especially in terms of detection time, substantiating the efficacy of our approach in real-world scenarios.