Efficient Machine Learning Research Articles

Developing an Efficient Support Vector Machine Learning Model for Mobile Charging Stations.

In the context of increasing global energy scarcity, optimizing electric vehicle (EV) mobile charging stations is critical for promoting sustainable transportation. This study introduces the use Artificial Neural Network (ANN) models, and Support Vector Machine (SVM) enhanced with the Adams optimizer, to address the challenge of efficient EV charging in energy-constrained environments. The models are designed to predict optimal charging station locations and schedules, with the ANN-Adams optimization fine-tuning the model parameters to improve accuracy and performance, while ensuring dynamic adaptation to fluctuating energy availability and demand patterns. This research contributes to the development of intelligent, adaptive systems for EV infrastructure, paving the way for more resilient and energy-efficient urban mobility solutions. The system employs supervised learning, where models were trained and tested on labeled datasets. The performance of the SVM models was compared to that of a Multilayer Perceptron Network (MLPN) with Adams optimization. The results showed that the MLPN with Adams optimization achieved an accuracy of 97.85%, while the SVM model had a prediction accuracy of 81%. The ANN recorded 91.29% accuracy. Both approaches significantly improved classification accuracy, model generalization on testing datasets, and reduced misclassification errors.

Predicting Loop Vectorization through Machine Learning Algorithms

Automatic vectorization is often utilized to improve the speed of compute-intensive programs on current CPUs. However, there is enormous space for improvement in present compiler auto-vectorization capabilities. Execution with optimizing code on these resource-controlled strategies is essential for both energy and performance efficiency. While vectorization suggests major performance developments, conventional compiler auto-vectorization techniques often fail. This study investigated the prospective of machine learning algorithms to enhance vectorization. The study proposes an ensemble learning method by employing Random Forest (RF), Feedforward Neural Network (FNN), and Support Vector Machine (SVM) algorithms to estimate the effectiveness of vectorization over Trimaran Single-Value Code (TSVC) loops. Unlike existing methods that depend on static program features, we leverage dynamic features removed from hardware counter-events to build efficient and robust machine learning models. Our approach aims to improve the performance of e-business microcontroller platforms while identifying profitable vectorization opportunities. We assess our method using a benchmark group of 155 loops with two commonly used compilers (GCC and Clang). The results demonstrated high accuracy in predicting vectorization benefits in e-business applications.

Evaluating the effectiveness of machine learning models for detecting AI-generated art

Breakthroughs in technology mean the creation of new programs that can produce results startlingly close to human creations. AI-generated art, for example, has garnered attention and debate by redefining artistic creations, originality, and the ethics of artificial intelligence. By training on existing artwork, these programs can generate images startlingly close to the intended art style and topic and can fool the ordinary person. In this paper, we examine whether the type and style of the input image will impact the accuracy of existing AI-detection machine learning models. The first part of the hypothesis is that the existing models are more accurate in classifying whether an image is AI or human-created if the image is centered around human subjects rather than the environment. The second part of the hypothesis is that the models are more accurate in classifying realistic images compared to images in animation style and traditional mediums. The data show that current models are slightly better at classifying images of the environment, with 84.7% accuracy, compared to images of human characters, with 81.9% accuracy, which doesn’t support the first part of our hypothesis. However, most models have greater success with realism style than with traditional and animation styles, which supports the second part of our hypothesis. These results may serve as suggestions for further improvements in current models. With efficient machine learning models, artists and the general public can discern between AI and human-created art, which may help improve the regulation and usage of AI-generated art.

Development of an efficient machine learning algorithm for reliable credit card fraud identification and protection systems

Recent developments in e-commerce and e-payment systems have led to a rise in financial fraud incidents, particularly credit card fraud. Software tools to identify credit card theft are essential. Critical characteristics of credit card fraud are crucial in utilizing Machine Learning (ML) for credit card fraud identification and must be selected carefully. This study suggests a An Efficient Machine Learning Algorithm for Reliable Credit Card Fraud Identification (EMLA-RCCFI) was constructed using ML, which utilizes the Genetic Algorithm (GA) to select features. Once the optimum characteristics are determined, the suggested detecting module utilizes the subsequent ML-based classifications. The proposed EMLA-RCCFI system is assessed using a dataset produced by European cardholders to confirm its efficacy. Based on the results, the suggested EMLA-RCCFI method surpassed existing systems regarding accuracy, precision, and F score.

Advancing pneumonia virus drug discovery with virtual screening: A cutting-edge fast and resource efficient machine learning framework for predictive analysis

Pneumonia, a severe respiratory infection characterized by a significant morbidity and fatality rate, afflicts many individuals globally. The demand for highly effective antiviral medications has experienced a surge because of the emergence of novel pneumonia viruses, such as the COVID-19 coronavirus. Due to their inherent time and cost constraints, conventional drug development strategies sometimes need to be more manageable. Exploring alternative approaches is crucial to identifying and establishing effective therapy choices. This work introduces a computational methodology for analyzing the chemical space of medications targeting the pneumonia virus, employing Python-based data mining tools. Using computer-aided analysis in drug molecules aims to enhance the efficiency of identifying and evaluating potential new therapeutic candidates using Machine Learning (ML). The research successfully discovered two therapeutic compounds by utilizing the Bayesian Ridge approach, which is the most accurate with the least mean squared error, is less computationally expensive in terms of power, memory, and CPU, and is the fastest of the investigated approaches. It discovered the CHEMBL433378 and CHEMBL93653, with promising docking scores of −4.3 and −4.2, respectively. Additionally, both molecules demonstrated significant inhibitory activity against their respective targets, as seen by their IC50 values of 0.0018 and 0.001. Both compounds meet the criteria for the B. Mann Whitney U Test and Lipinski test.

Highly Efficient Machine Learning Approach for Automatic Disease and Color Classification of Olive Fruits

Highly Efficient Machine Learning Approach for Automatic Disease and Color Classification of Olive Fruits

Efficient machine learning models for the detection of coconut milk adulteration

Coconut milk adulteration detection involves the use of analytical methods, including machine learning, to detect the presence of impurities or additives in coconut milk. Adulteration can occur when substances such as water or other cheaper ingredients are added to coconut milk, compromising its quality, nutritional value, and authenticity. Identifying adulteration is crucial for ensuring consumer safety, maintaining product quality, and upholding industry standards. The aim of this study was to propose a machine learning model to detect adulteration in coconut milk. To implement the proposed work, a coconut milk adulteration dataset is collected from a standard source. The amount of available data is limited; hence, synthetic data is generated by applying the CTGAN algorithm. The proposed framework employs three different Feature Extraction (FE) strategies, i.e., Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), and Autoencoder (AE). Then, the feature-extracted dataset is classified through four effective machine learning algorithms, such as Logistic Regression (LR), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF). The machine learning algorithms outcomes are evaluated using four performance metrics, i.e., Accuracy, Precision, Recall, and F1-score. RF provides the highest accuracy of 99.17%, precision value of 97.29%, recall value of 96.71%, and F1 Score of 97% for the LDA techniques.

Element-configuration dependent first-principles machine learning studies of multiple alloying effects on the structure stability of Co3(Al, W)

Multiple alloying is desired to improve the stability of metastable Co3(Al, W) of Co-based superalloys. However, computational studies of alloying effects are costly due to the enormous combinations of compositions and configurations. Most current element-based machine learning methods that rely solely on composition lack the ability to distinguish between polymorph materials and various local alloying configurations. In this work, we developed efficient machine learning (ML) models using element-configuration dependent Center-Environment (CE) and element-based Chemical Composition (CC) features to describe alloying element type and substitution configuration, crucial for data-driven study of correlated alloying effects and local short-range chemical ordering. We employed first-principles computation and ML methods to investigate the structure stabilities of the Al-W substitution site models with 3d and 4d transition metal (TM) elements, then the optimized ML models were validated using the untrained dual-Al or dual-W site configurations with 3d, 4d, and 5d TM elements. The ML-CE models were generally more accurate than the ML-CC models. The most stable substitution pairs are Sc-Fe 3d TM elements at the Al-W and dual-W sites with lattice constant ∼4 Å. Therefore, we propose a “Co-left” design rule to stabilize Co3(Al, W) by multiple alloying with early 3d TM elements (IIIB-VIII group). The stabilizing 3d “Co-left” elements also can improve the mechanical properties. The ML-CE approach can be generally applied to study local correlated alloying effects and chemical ordering in medium/high-entropy or amorphous alloys.

Employee benefits and company performance: Evidence from a high-dimensional machine learning model

By incorporating novel social media data, we analyze in detail how US companies offer different employee benefits and how they are associated with several company performance measures. Benefits such as 401(k), employee discounts, parking, and vision/dental healthcare are the most commonly provided, while free food -related benefits and family-related benefits are the most scarcely offered. Furthermore, with the aid of efficient machine learning -based models and tools from explainable artificial intelligence, we discover that family-related benefits are often associated with the most satisfied employees and best-performing companies. Our findings indicate that high-growth companies tend to provide a broad array of benefits to their employees. In contrast, highly profitable companies often concentrate on delivering a more limited and specialized set of benefits. We argue that companies offer rare and highly sought benefits to keep and recruit high-performers.

Efficient machine learning model to detect early stage Parkinson’s disease

<p class="root-block-node">Parkinson’s disease (PD) often manifests itself in memory loss and cognitive decline. The decline is inexorable, and damage to the brain’s cortex has already occurred. Numerous studies have shown that by detecting dementia early and beginning treatment, the disease’s course can be slowed, and any further atrophy can be prevented. Brain imaging data, such as from an MRI, is frequently used in the diagnosis of Parkinson’s disease (PD). In recent years, utilizing deep convolutional neural networks has greatly improved Parkinson’s disease diagnosis. However, getting to the level of quality needed for clinical use is still challenging. In this study, we introduce a machine learning-based approach for more accurately diagnosing Parkinson’s disease. This research makes use of information gleaned from single-photon emission computerized tomography (SPECT) scan and positron emission tomography (PET) scans performed on patients with Parkinson’s disease (PD) and healthy controls. The most crucial characteristics of these datasets are isolated with the aid of the Fisher discriminate ratio (FDR) and non-negative matrix factorization (NMF). The K-nearest neighbor, Decision Tree, Support vector machine (SVM), and Deep Convolution neural network (DCCN) classifiers with confidence bounds classify the NMF-transformed data sets with a decreased number of features. The proposed DCCN technique has a classification accuracy of up to 93.7 percent when compared to decision trees, K-Nearest Neighbor (KNN)s, and SVMs. The DCCN is now a reliable approach for classifying SPECT and PET, PD images.</p>

Photonic reservoir computing for energy efficient and versatile machine learning application

Time-multiplexed reservoir computing is a machine learning concept which can be realised in photonic hardware systems using only one physical node. The concept can be used for various problems, ranging from classification problems to time-series prediction tasks, while being fast and energy efficient. Here, a theoretical analysis of a reservoir computer realised via delay-coupled semiconductor lasers is presented and the role of the internal system time-scales and the bifurcation structure is discussed. It is further shown that optimal performance can be reached by tailoring the coupling delays to the specific memory requirements of the given task.

Credit Default Analysis and Prediction Based on Machine Learning

The Credit Default Prediction project aims to develop an efficient machine learning model to accurately predict loan default risk on the Lending Club platform. The project is based on historical loan data, including borrower personal information, financial metrics, and credit history records, with the goal of building a robust predictive model. The project's workflow encompasses several critical steps, including data preprocessing, exploratory data analysis, feature engineering, model selection, and performance evaluation. In terms of model selection, the project employs two primary machine learning algorithms, namely Artificial Neural Networks (ANN) and Random Forest. These algorithms are renowned for their outstanding performance in handling extensive borrower data and providing reliable risk predictions. Model training and evaluation are conducted using a substantial amount of historical data to ensure accurate predictions across various scenarios. Furthermore, the project conducts feature importance analysis to identify factors that significantly influence loan default risk. These insights contribute to enhancing Lending Club's risk assessment process and supporting more informed decisions regarding loan approval and pricing strategies. By combining data-driven predictive modeling with in-depth data analysis, this project aims to enhance the efficiency of Lending Club's loan operations and elevate its risk management capabilities, ultimately providing more dependable financial services to investors and borrowers.

Visualization and characterization of agricultural sprays using machine learning based digital inline holography

Accurate characterization of agricultural sprays is crucial to predict in-field performance of liquid applied crop protection products. In this paper, we introduce a robust and efficient machine learning (ML) based Digital In-line Holography (DIH) algorithm to accurately characterize the droplet field for a wide range of commonly used agricultural spray nozzles. Compared to non-ML based DIH processing, the ML-based algorithm enhances accuracy, generalizability, and processing speed. The ML-based approach employs two neural networks: a modified U-Net to obtain the 3D droplet field from the numerically reconstructed optical field, followed by a VGG16 classifier to reduce false positives from the U-Net prediction. The modified U-Net is trained using holograms generated using a single spray nozzle (XR11003) at three different spray locations; center, half-span, and the spray edge to create training data with different number densities and droplet size ranges. VGG16 is trained using the minimum intensity projection of the droplet 3D point spread function (PSF). Data augmentation is used to increase the efficiency of classification and make the algorithm generalizable for different measurement settings. The model is validated using National Institute of Standards and Technology (NIST) traceable glass beads and six different agricultural spray nozzles representing various spray characteristics. The principal results demonstrate a high accuracy rate, with over 90% droplet extraction and less than 5% false positives, regardless of droplet number density and size. Compared to traditional spray measurement techniques, the new ML-based DIH methodology offers a significant leap forward in spatial resolution and generalizability. We show that the proposed ML-based algorithm can extract the real cumulative volume distribution of the NIST beads, where the LD system measurements are biased towards droplets moving at slower speeds. Additionally, the ML-based DIH approach enables the estimation of mass and momentum flux at different locations and the calculation of relative velocities of droplet pairs, which are difficult to obtain using conventional spray characterization techniques.

Overview of Big Data Analytics in Modern Astronomy

Astronomers are increasingly compelled to chart the universe with ever greater precision. Projects like the Sloan Digital Sky Survey (SDSS), Pan-STARRS, and the Large Synoptic Survey Telescope (LSST) generate approximately 100-200 Petabytes of data annually, presenting significant big data challenges in terms of storage, processing, and data transfer. The Square Kilometer Array (SKA), an ambitious project involving 130,000 antennas and 200 dishes spanning two continents, is scheduled to become operational in 2028. It will collect 160 terabytes of data per second, translating to 1 petabyte of data daily. Coping with this immense volume of data necessitates real-time processing and analysis, driving the need for efficient machine learning and image analysis algorithms. Astronomy stands as an ideal domain for big data analytics, pushing the boundaries of data analysis. This review paper will present intriguing applications for data scientists, exploring the challenges and recent technological advancements in big data analytics concerning astronomy. The paper will also critically assess the strengths and weaknesses of various approaches, methodologies, or tools used in big data analytics within the context of astronomy, supported by relevant case studies.

Towards Reliable Federated Learning Using Blockchain-Based Reverse Auctions and Reputation Incentives

In recent years, the explosion of big data has presented unparalleled opportunities for the advancement of machine learning (ML). However, the vast size and sensitive nature of these datasets present significant challenges in terms of privacy and security. Federated Learning has emerged as a promising solution that enables a group of participants to train ML models without compromising the confidentiality of their raw data. Despite its potential, traditional federated learning faces challenges such as the absence of participant incentives and audit mechanisms. Furthermore, these challenges become more significant when dealing with the scale and diversity of big data, making efficient and reliable federated learning a complex task. These limitations may compromise model quality due to potential malicious nodes. To address the above issues, this paper proposes a BlockChain-based Decentralized Federated Learning (BCD-FL) model. In BCD-FL, we design a smart contract approach based on the reverse auction-based incentive mechanism and a reputation mechanism to promote the participation of reliable and high-quality data owners. Theoretical analysis shows that the BCD-FL model satisfies several desirable properties, such as individual rationality, computational efficiency, budget balance, and truthfulness. In addition, experimental results also show that the proposed model enables more efficient federated learning and provides some level of protection against malicious nodes. Therefore, the BCD-FL model presents a potential solution to the challenges in federated learning and opens up new possibilities for achieving efficient large-scale machine learning.

Prediction of Parkinson's Disease Using Machine Learning Methods.

The detection of Parkinson's disease (PD) in its early stages is of great importance for its treatment and management, but consensus is lacking on what information is necessary and what models should be used to best predict PD risk. In our study, we first grouped PD-associated factors based on their cost and accessibility, and then gradually incorporated them into risk predictions, which were built using eight commonly used machine learning models to allow for comprehensive assessment. Finally, the Shapley Additive Explanations (SHAP) method was used to investigate the contributions of each factor. We found that models built with demographic variables, hospital admission examinations, clinical assessment, and polygenic risk score achieved the best prediction performance, and the inclusion of invasive biomarkers could not further enhance its accuracy. Among the eight machine learning models considered, penalized logistic regression and XGBoost were the most accurate algorithms for assessing PD risk, with penalized logistic regression achieving an area under the curve of 0.94 and a Brier score of 0.08. Olfactory function and polygenic risk scores were the most important predictors for PD risk. Our research has offered a practical framework for PD risk assessment, where necessary information and efficient machine learning tools were highlighted.

Crack localization in glass fiber composite beams by experimental modal analysis and multi variable Gaussian process regression method

PurposeEnsuring the early detection of structural issues in aircraft is crucial for preserving human lives. One effective approach involves identifying cracks in composite structures. This paper employs experimental modal analysis and a multi-variable Gaussian process regression method to detect and locate cracks in glass fiber composite beams.Design/methodology/approachThe present study proposes Gaussian process regression model trained by the first three natural frequencies determined experimentally using a roving impact hammer method with crystal four-channel analyzer, uniaxial accelerometer and experimental modal analysis software. The first three natural frequencies of the cracked composite beams obtained from experimental modal analysis are used to train a multi-variable Gaussian process regression model for crack localization. Radial basis function is used as a kernel function, and hyperparameters are optimized using the negative log marginal likelihood function. Bayesian conditional probability likelihood function is used to estimate the mean and variance for crack localization in composite structures.FindingsThe efficiency of Gaussian process regression is improved in the present work with the normalization of input data. The fitted Gaussian process regression model validates with experimental modal analysis for crack localization in composite structures. The discrepancy between predicted and measured values is 1.8%, indicating strong agreement between the experimental modal analysis and Gaussian process regression methods. Compared to other recent methods in the literature, this approach significantly improves efficiency and reduces error from 18.4% to 1.8%. Gaussian process regression is an efficient machine learning algorithm for crack localization in composite structures.Originality/valueThe experimental modal analysis results are first utilized for crack localization in cracked composite structures. Additionally, the input data are normalized and employed in a machine learning algorithm, such as the multi-variable Gaussian process regression method, to efficiently determine the crack location in these structures.

Failure Probability Prediction for Offshore Floating Structures Using Machine Learning

Summary Accurately estimating the failure probability is crucial in designing civil infrastructure systems, such as floating offshore platforms for oil and gas processing/production, to ensure their safe operation throughout their service periods. However, as a system becomes complex, the evaluation of a limit state function may involve the use of an external computer solver, resulting in a significant computational burden to perform Monte Carlo simulations (MCS). Moreover, the high-dimensionality of the limit state function may limit efficient sampling of input variables due to the “curse of dimensionality.” To address these issues, an efficient machine learning framework is proposed, combining polynomial chaos expansion (PCE) and active subspace. This will enable the accurate and efficient evaluation of the failure probability of an offshore structure, which typically involves a large number of uncertain parameters. Unlike conventional PCE schemes that use the original random variable space or the auxiliary variable space for building a surrogate model, the proposed method utilizes a reduced-dimension space to circumvent the “curse of dimensionality.” An appropriate coordinate transformation is first sought so that most of the variability of a limit state function can be accounted for. Next, a PCE surrogate limit state function is constructed on the derived low-dimensional “active subspace.” The Gram-Schmidt orthogonalization process is used for making basis polynomial functions, which is particularly effective when input random parameters do not follow the Askey scheme and/or when a dependence structure between the input parameters exists. Therefore, a nonlinear iso-probabilistic transformation, which makes the convergence of a surrogate to the true model difficult, is not required, unlike traditional PCE. Numerical examples, including limit state functions related to structural dynamics problems, are presented to illustrate the advantages of the proposed method in estimating failure probabilities for complex structural systems. Specifically, the method exhibits significantly improved efficiency in estimating the failure probability of an offshore floating structure without compromising accuracy as compared to traditional PCE and MCS.

Plant Leaf Disease Detection Using Efficient Image Processing and Machine Learning Algorithms

India is often described as a country of villages, where a majority of the population depends on agriculture for their livelihood. The landscape of Indian agriculture is approximately 159.7 million hectares. Agriculture plays a pivotal role in India's Gross Domestic Product (GDP), accounting for about 18% of the nation's economic output. Diseases and pests can have detrimental effects on crops, leading to reduced yields. These challenges can include the spread of plant diseases, infestations by insects or other pests, and the overall degradation of crop health. Early detection of diseases in crops is crucial for several reasons. Detecting diseases at an early stage allows for prompt intervention, such as applying appropriate pesticides or taking preventive measures. The main aim of this study is to develop a highly effective method for plant leaf disease detection using computer vision techniques. Here, leaf disease detection comprises histogram equalization, denoising, image color threshold masking, feature descriptors such as Haralick textures, Hu moments, and color histograms to extract the salient features of leaf images. These features are then used to classify the images by training Logistic Regression, Linear Discriminant Analysis, K-nearest neighbor, decision tree, Random Forest, and Support Vector Machine algorithms using K-fold validation. K-fold validation is used to separate the validation samples from the training samples, and the K indicates the number of times this is repeated for the generalization. The training and validation processes are performed in two approaches. The first approach uses default hyperparameters with segmented and non-segmented images. In the second approach, all hyperparameters of the models are optimized to train segmented datasets. The classification accuracy improved by 2.19% by utilizing segmentation and hyperparameter tuning further improved by 0.48%. The highest average classification accuracy of 97.92% is achieved using the Random Forest classifier to classify 40 classes of 10 different plant species. Accurate detection of plant disease leads to the sustained growth of plants throughout the growing span of the plants.

Real-time yaw-misalignment calibration and field-test verification of wind turbine via machine learning methods

It has become a general consensus that nacelle-mounted LiDAR can be used to calibrate the yaw misalignment or drive the real-time yaw motions for wind turbines, which would improve the power-generation efficiency. The advantage of LiDAR utilization is that the accuracy of inflow wind measurement would be greatly improved, while its disadvantage is that the cost remains high and the data validity is not sufficiently high. In this paper, an efficient machine learning method for estimating LiDAR measurement is developed to establish the real-time yaw calibration framework and sustain the LiDAR rolling utilization. Firstly, the correlation of LiDAR measurement with SCADA features is analyzed to estimate LiDAR measurement using only SCADA data. Secondly, several machine learning algorithms are studied for performance comparison, and the dependence of each algorithm on data size is also analyzed. Experimental results show that, the proposed XGBoost algorithm has high accuracy, requires less data, and can quickly calibrate the yaw misalignment. Finally, the field testing is held for a commercial 2 MW wind turbine to verify the effectiveness. The field-test results show that the proposed method is feasible for industrial applications and can improve the annual theoretical power generation by 3.66% compared to the situation without calibration, which also provides an executable and economical solution for LiDAR replacement planning.