Data Machine Learning Research Articles

A Comprehensive Review of the Evaluation and Application of Engineering Properties of Expansive Soil Based on Piezocone Penetration Test (CPTU)

Expansive soil presents several challenges in civil and geotechnical engineering because of its distinctive physical and mechanical properties, including considerable volumetric fluctuations, non-linear stress–strain behaviour, and high permeability. Owing to soil disturbances and sample size limitations during sampling, important parameters such as the expansive soil expansion force, compressive strength, and shear strength obtained through indoor experiments often cannot accurately reflect the values in situ. In this regard, in situ piezocone penetration testing has the advantages of small disturbances and high efficiency and has been widely used in subways, bridges, highways, and other engineering projects. This paper comprehensively summarizes the research progress of in situ testing technology in the evaluation and application of engineering properties of expansive soils in China and internationally. First, a detailed analysis of the basic theories and physical mechanisms of various in situ testing methods is provided, focusing on their applicability and accuracy in expansive soils. Second, examples of the application of in situ testing technology in expansive soil engineering are summarized. Finally, predictive models and methods based on in situ testing data, especially those that introduce cutting-edge models such as machine learning and advanced data analysis techniques, are discussed.

Machine learning model for deep exploration: Utilizing short wavelength infrared (SWIR) of hydrothermal alteration minerals in the Qianchen gold deposit, Jiaodong Peninsula, Eastern China

The Jiaodong Peninsula in the eastern North China Craton is renowned globally as a major gold province. Notably, the Jiaojia orefield is an important gold-producing district in this region, and its depths are inferred to offer substantial prospecting potential, underscoring the need for innovative exploration tools to uncover concealed orebodies. The Qianchen gold deposit in the southern part of the Jiaojia orefield is characterized by the disseminated-stockwork-style orebodies at depths of ca. –1000 to –1800 m, with three mineralization stages, namely (I) pre-ore K-feldspar-quartz stage, (II) quartz-sericite-gold-polymetallic sulfide stage (main Au mineralization stage), and (III) post-ore quartz-calcite stage. Systematic short-wavelength infrared (SWIR) spectral analysis reveals that sericite group minerals developing in Stage II are the predominant and wide-distributed alteration minerals at Qianchen. Their spectral characteristics demonstrate a shift in the Al-OH absorption feature wavelength positions (Pos2200) to longer wavelengths (>2205 nm) within the orebodies. This shift is likely due to strong water–rock interactions leading to Fe and Mg substituting for Al in octahedral sites. Orthogonal partial least squares discriminant analysis (OPLS-DA) indicates that the H2O, OH, and FeOH spectral bands (Variable Importance in the Projection (VIP) ≥ 1) of sericite are crucial for distinguishing between gold ores and barren wall rocks. Additionally, Long Short-Term Memory (LSTM) models demonstrate that spectral data of sericite can be used to predict ore samples with an accuracy of 82 %, validated by receiver operating characteristic analysis (84 %), and blind test results demonstrate that the predicted grade closely aligns with the actual grade. Therefore, it is concluded that SWIR spectral characteristics of sericite group minerals can serve as an effective indicator for prospecting the Qianchen concealed gold mineralization. This study highlights the potential of coupling between machine learning and SWIR spectral data of alteration minerals to uncover the concealed lode gold mineralization.

Preface

The 2023 International Conference on Applied Mathematics and Digital Simulation (AMDS 2023) was held successfully on 28-29 December 2023 in Beijing, China. The contributions are the results of submissions from the conference participants, who provide readers with a broader view of mathematical, statistical and computational science methods, ideas and tools.There are 92 participants in AMDS 2023, in which the program includes 6 keynote reports (35-45 minutes), 48 oral presentations (15 mins+3 Q&A) included in four sessions (Mathematics and Computational Theory, Mathematical Modelling and Machine Learning Methods Statistics Modelling and Data Analysis, and Fluid Dynamics and Mechanics), 15 E-posts and 23 listeners. The point of the conference emphasizes the development of theories and applications and promotes interdisciplinary endeavors among mathematicians, statisticians, scientists, engineers and researchers from other related disciplines.This volume comprises selected papers from the AMDS2023 conference. Accepted manuscripts were peer-reviewed by 2 to 4 reviewers in a double-blind review process. A total of 48 reviewers participated in this conference and provided their professional comments to authors to ensure the quality of the manuscripts. There are six topic partitions, including Mathematics and Computational Theory, Statistics Modelling and Data Analysis, Mathematical Modelling and Machine Learning Methods, Fluid Dynamics and Mechanics, Mathematical Modelling in Physics and Control and Automation, which constitutes an up-to-date account of principles, techniques, and tools for mathematical simulation and modelling in a wide range of research fields.We would like to thank the authors of all papers and participants; the members of the Scientific Committee who assisted with the review of the papers submitted and presented at the conference; the four session chairs: Prof. Yichuan Zhao, Prof. Quanxin Zhu, Prof. P.M. Siva Raja, Prof. Rao Li; the six keynote speakers: Prof. Alexander G. Ramm, Prof. Ovidiu Radulescu, Prof. Hajime Urakawa, Prof. Yanjun Liu, Prof. Grienggrai Rajchakit and Prof. Oscar Eduardo Ruiz Salguero; the organzier and sponsors: the Hubei Mengshun Exhibition Service Center, the Research Center for Stochastic Configuration Machines (SCM), the National Laboratory for Scientific Computing (LNCC); the publisher: IOP Publishing Ltd agrees to publish the AMDS 2023 proceedings. We are grateful for your attendance and contribution to a successful AMDS 2023.Organizing Cmmittee of AMDS 2023Jiang Zhu (Advisory Chair)National Laboratory for Scientific Computing-LNCC, BraziList of Conference Committees are available in this Pdf.

ANIMAL GAZING DETECTION USING CNN IN SMART AGRICULTURAL FIELD

Animal detection and warning systems play a critical role in safeguarding agricultural land from the intrusion of large animals, particularly ungulates, which pose threats to property and human safety. These systems are designed to identify and alert the presence of such animals before they encroach upon farming territories. Across different regions, the deployment of these systems aims to mitigate Animal Vehicle Collisions (AVCs) involving ungulate species, which can result in significant costs and hazards. However, the effectiveness of these systems may be hindered by the lack of regular monitoring for incidents of animal road fatalities. This paper explores the operational principles of animal detection systems, highlighting their reliance on machine learning technology to detect patterns indicative of animal presence. The detection process involves analyzing data collected from various sources to discern behavioral and environmental cues associated with animal activity. Challenges may arise from accurately distinguishing between different types of animals and background elements, necessitating the refinement of machine learning algorithms to minimize false positives triggered by environmental factors. Effective deployment of these systems requires comprehensive planning and integration into regional and national agricultural policies. By leveraging advancements in machine learning technology and data analytics, stakeholders can enhance the efficiency of these systems, thereby reducing the incidence of animal- related accidents and protecting agricultural resources. This paper underscores the importance of continuous monitoring and adaptation of animal detection systems to effectively mitigate risks and ensure the sustained protection of farming land and communities.

Deep Learning Models for Type 2 Diabetes Detection in Saudi Arabia

One of the predominant health issues affecting Saudi Arabia and leading to many complications is Type 2 diabetes (T2D). Early detection and significant preventative measures lead to curbing and controlling the health issue. There are fewer datasets in the literature for the detection of T2D in the Saudi population. Past studies using Saudi data have favoured machine learning algorithms to classify T2D. Although the application of this data in machine learning is evident, no studies exist in the literature that compare this data, especially those related to deep learning algorithms. This study's objective is to use specific Saudi data to develop multiple deep learning models that could be used to detect T2D. The research uses a Deep Neural Network (DNN), an Autoencoder (AE), and a Convolutional Neural Network (CNN) to create predictive models and compare their performance with a traditional machine learning classifier used on the same dataset that outperformed other machine learning algorithms such as a Decision Forest (DF). Various metrics were used to evaluate the effectiveness of the models, such as accuracy, precision, recall, F1 score and area under the ROC curve (AUC) where the ROC acts as a receiver operating characteristic curve. There are two cases in this paper: (i) uses all features of the dataset and (ii) uses six of the ten features, such as DF. In case (i), the results were shown that AE outperformed other models with the highest accuracy for imbalanced and balanced data 81.12\% and 79.16\%, respectively. The results for case (ii) showed that AE scored the highest 81.01\% accuracy with imbalanced data compared to DF and DF achieved the highest accuracy of 82.1\% with balanced data. As a result, both cases explored in this study revealed that AE has a constant superior performance if imbalanced data is used. In contrast, DF demonstrated the highest accuracy when a balanced dataset was used with a feature set reduction. They help to identify the undiagnosed T2D, and they are essential for professionals in Saudi Arabia in the health sector to promote health connections, identify risks and contain or improve their diabetes management.

Enhancing Study Space Utilization at UCL: Leveraging IoT Data and Machine Learning

University College London (UCL) has more than 51,000 students along with a challenge to find available study space. UCL has improved its Internet of Things (IoT) infrastructure to better analyze space occupancy in order to solve this. Current systems offer real-time information, but they are only available in crowded places. Therefore, this study presents a data analysis system that recommends the best study areas by using Machine Learning (ML) and historical data. The main library, student center, and science library at UCL are the top three study places. The goal of this study is to transform raw occupancy data from these venues into useful insights. The development of an independent system written in Python was facilitated by research into dataset optimization, ML regressions, popularity patterns, and feature analysis. This technique uses historical trends to forecast occupancy rates for the future.

Hardness-guided machine learning for tungsten alloy strength prediction

Yield strength determines the material's resistance to permanent deformation. However, the traditional estimation method based on the triple relationship of Vickers hardness exhibits limited accuracy when applied to tungsten heavy alloys (WHA) with two-phase structure. This study presents a novel approach using machine learning and hardness data to predict the yield strength of tungsten heavy alloy efficiently and accurately. Among the eight machine learning models used, Gradient boosting decision tree model (GBDT) was the most effective model. The results demonstrate a strong agreement between the GBDT-based predictions and experimental results, with an average error of 6.7 %. We also gave the hardness versus yield strength relationship for the 90WNiFe alloy system. This finding highlights the potential of the GBDT regression model in conjunction with hardness measurements for predicting yield strength with high precision.

Personalized Recommendation of English Online Teaching Content Based on Logistic Regression Algorithm

English online teaching content encompasses a wide range of materials designed to facilitate language learning in digital environments. These resources include interactive lessons, video tutorials, audio recordings, quizzes, vocabulary exercises, grammar explanations, and authentic texts such as articles, short stories, and dialogues. The content aims to cover various language skills, including listening, speaking, reading, and writing, catering to learners of different proficiency levels and learning styles. This paper introduces a Cross-Domain Personalized Recommendation System (CDPRS) designed to enhance the delivery of English language teaching content. Leveraging machine learning algorithms and data analytics techniques, the CDPRS tailors recommendations to individual user preferences, learning objectives, and proficiency levels across diverse content domains. Through a series of analyses and evaluations, including regression analysis and simulation studies, the effectiveness and utility of the CDPRS are demonstrated. Through a series of analyses and evaluations, including regression analysis and simulation studies, the effectiveness and utility of the CDPRS are demonstrated. Results indicate that the system achieves high levels of accuracy (mean accuracy = 0.85), diversity (mean diversity = 0.72), novelty (mean novelty = 0.78), and user satisfaction (mean satisfaction = 0.82) in providing personalized recommendations.

Machine Learning based Data Sensing Device Network with Optimized Group Formation and Group Head Selection

Sensor devices for data sensing (referred to as DSDs) are used in use cases such as border control and vehicle tracking. The architecture of the Data Sensing Device Network (DSDN) is established by integrating numerous DSDs across a given region, forming multiple groups. Within each group, a specific DSD is designated to facilitate communication between independent groups. The multi attribute values are captured and these attribute values effect the selection of head DSDs. For each of DSD this value ranges between 0.1 to 1. The DSD which has the highest value of range will be treated as Group Head in LEACH. The attributes are namely distance, battery level for each DSD. From the source DSD to destination DSD the link formation will happen end to end by making use of DSDs and base station, generally the end-to-end link communication has larger hops. This will have a ripple effect on battery level for DSDs and can cause reduced lifetime. The Energy based LEACH is modified on top of LEACH by computing the battery level for DSDs and picking DSD with highest battery level. The Energy based LEACH will have two DSDs in each group acting like head DSDs. Machine Learning Data Sensing Device Network (ML-DSDN) is proposed which will first create group DSDNs based on k means machine learning algorithm. ML-DSDN will find the head group DSD based on combination of random forest and SVM algorithm with set theory. The comparison is done of ML-DSDN with respect to ELEACH and LEACH method and it is proved that ML-DSDN performs better with respect to delay, link count, energy consumption, alive DSD count, dead DSD count, lifetime ratio, routing overhead.

Machine Learning Data Augmentation Strategy for Electron Energy Loss Spectroscopy: Generative Adversarial Networks.

Recent advances in machine learning (ML) have highlighted a novel challenge concerning the quality and quantity of data required to effectively train algorithms in supervised ML procedures. This article introduces a data augmentation (DA) strategy for electron energy loss spectroscopy (EELS) data, employing generative adversarial networks (GANs). We present an innovative approach, called the data augmentation generative adversarial network (DAG), which facilitates data generation from a very limited number of spectra, around 100. Throughout this study, we explore the optimal configuration for GANs to produce realistic spectra. Notably, our DAG generates realistic spectra, and the spectra produced by the generator are successfully used in real-world applications to train classifiers based on artificial neural networks (ANNs) and support vector machines (SVMs) that have been successful in classifying experimental EEL spectra.

A machine learning screening model for identifying the risk of high-frequency hearing impairment in a general population

BackgroundHearing impairment (HI) has become a major public health issue in China. Currently, due to the limitations of primary health care, the gold standard for HI diagnosis (pure-tone hearing test) is not suitable for large-scale use in community settings. Therefore, the purpose of this study was to develop a cost-effective HI screening model for the general population using machine learning (ML) methods and data gathered from community-based scenarios, aiming to help improve the hearing-related health outcomes of community residents.MethodsThis study recruited 3371 community residents from 7 health centres in Zhejiang, China. Sixty-eight indicators derived from questionnaire surveys and routine haematological tests were delivered and used for modelling. Seven commonly used ML models (the naive Bayes (NB), K-nearest neighbours (KNN), support vector machine (SVM), random forest (RF), eXtreme Gradient Boosting (XGBoost), boosting, and least absolute shrinkage and selection operator (LASSO regression)) were adopted and compared to develop the final high-frequency hearing impairment (HFHI) screening model for community residents. The model was constructed with a nomogram to obtain the risk score of the probability of individuals suffering from HFHI. According to the risk score, the population was divided into three risk stratifications (low, medium and high) and the risk factor characteristics of each dimension under different risk stratifications were identified.ResultsAmong all the algorithms used, the LASSO-based model achieved the best performance on the validation set by attaining an area under the curve (AUC) of 0.868 (95% confidence interval (CI): 0.847–0.889) and reaching precision, specificity and F-score values all greater than 80%. Five demographic indicators, 7 disease-related features, 5 behavioural factors, 2 environmental exposures, 2 hearing cognitive factors, and 13 blood test indicators were identified in the final screening model. A total of 91.42% (1235/1129) of the subjects in the high-risk group were confirmed to have HI by audiometry, which was 3.99 times greater than that in the low-risk group (22.91%, 301/1314). The high-risk population was mainly characterized as older, low-income and low-educated males, especially those with multiple chronic conditions, noise exposure, poor lifestyle, abnormal blood indices (e.g., red cell distribution width (RDW) and platelet distribution width (PDW)) and liver function indicators (e.g., triglyceride (TG), indirect bilirubin (IBIL), aspartate aminotransferase (AST) and low-density lipoprotein (LDL)). An HFHI nomogram was further generated to improve the operability of the screening model for community applications.ConclusionsThe HFHI risk screening model developed based on ML algorithms can more accurately identify residents with HFHI by categorizing them into the high-risk groups, which can further help to identify modifiable and immutable risk factors for residents at high risk of HI and promote their personalized HI prevention or intervention.

Predicting wetland soil properties using machine learning, geophysics, and soil measurement data

PurposeMachine learning models can improve the prediction of spatial variation of wetland soil properties, such as soil moisture content (SMC) and soil organic matter (SOM). Their performance, however, relies on the quantity of data used to train the model, limiting their use with insufficient data. In this study, we assessed the use of synthetic data constrained by limited field data for training an eXtreme Gradient Boosting (XGBoost) algorithm used to predict the distribution of soil properties based on geophysical measurements constrained by soil samples.Materials and methodsA spatial distribution of soil apparent electrical conductivity (ECa) and laboratory measurements of SOM and SMC from twenty-two core samples were acquired at the St. Michael restored wetland near Defiance, Ohio. The correlations between ECa, SOM, and SMC were explored for predicting the spatial distribution of SOM and SMC. We used a Beta Variational AutoEncoder (β-VAE) approach to synthetically generate over 70,000 training data from the original twenty-two data from soil cores. The training data samples were taken from the latent space. The XGBoost algorithm was then trained on the β-VAE generated data and used to predict the spatial distribution of SOM and SMC at the site. We also validated the accuracy of the XGBoost predictions using an original holdout model validation technique.Results and discussionsThe generated synthetic data using the β-VAE include both soil attributes and ECa, which are larger and more diverse than the original training set with an absolute mean reconstructed error for SMC and SOM ranging from 0.018 to 0.022 and 0.026 to 0.041, respectively. This indicates that the β-VAE successfully generated a realistic synthetic dataset and overcame the technical barrier of using limited datasets. In addition, using generated data to expand the original training data helps the XGBoost model make more accurate predictions compared to training on the original data. The XGBoost prediction performance yielded average Lin’s concordance correlation coefficient (LCCC) values of 0.82 and 0.85 for SOM and SMC and a ratio of performance to deviation (RPD) values of 1.92 and 2.22 respectively, indicating a good performance.ConclusionsThis study validated the use of β-VAE to successfully generate synthetic wetland soil datasets with attributes of the original field data that can be effectively used to train the machine learning XGBoost model. The proposed framework offers an efficient solution for mapping the spatial variability of soil property in data-scarce wetland soil environments.

Bridging machine learning and diagnostics of the ESA LISA space mission with equation discovery via explainable artificial intelligence

The Laser Interferometer Space Antenna (LISA) of the European Space Agency will be the first interferometer for low-frequency (10-4–10-1 Hz) gravitational wave detection in space. LISA will detect spurious accelerations of the test masses, the end mirrors of the interferometer, of the order of femto-g. Amongst spurious signals, Coulomb forces due to stray electric fields coupling with charge deposited on the test masses by galactic and solar particles and the noise associated to the charging process must be monitored during the mission lifetime. Precious clues on spurious forces acting on the test masses have been studied with the LISA Pathfinder mission, meant for the testing of the instrumentation that will be placed on board LISA, in 2016–2017. In this work we present the design and implementation of a workflow leading to equation discovery for the relationship between solar wind speed, galactic cosmic-ray variations and interplanetary magnetic field intensity observations for the diagnostics of LISA. The workflow exploits explainable artificial intelligence tools to build a bridge between the opaque predictions obtained with machine learning (ML) models and data analysis performed by humans with space mission observations. The core step of the workflow is the implementation, tuning and optimisation of an opaque ML regressor explicitly designed for the future LISA mission, based on input observations of the galactic cosmic-ray flux and of the interplanetary magnetic field intensity. The regressor is aimed at reconstructing the solar wind speed, a parameter of fundamental importance, but not available, for the mission environment monitoring. The workflow ends with the application of explainable clustering techniques to the opaque model in order to (i) obtain human-interpretable outputs instead of opaque ones without any noticeable loss in the predictive performance and (ii) discover an equation describing the quantitative relationship between the involved variables, currently missing in the literature. The correlation amongst the transit of interplanetary structures, galactic cosmic-ray variations and LISA test-mass charging is illustrated here.

АСПЕКТИ ВИКОРИСТАННЯ ШТУЧНОГО ІНТЕЛЕКТУ В СУЧАСНИХ ІНФОРМАЦІЙНИХ ТЕХНОЛОГІЯХ І УПРАВЛІННІ ВИРОБНИЦТВОМ

The work explores the possibilities of modern information technology in the organization of production using artificial intelligence to achieve high efficiency and increase competitiveness. The application of machine learning algorithms and data analysis allows you to increase productivity, reduce costs and improve production management. Artificial intelligence is used in product quality management by analyzing data on production processes, detecting defects and predicting possible problems. The main problems of introducing artificial intelligence into computer systems are the impossibility of predicting all real situations and programming the machine to respond appropriately to them, insufficient reliability and errors in the software. Legal issue of liability for errors of artificial intelligence.

Integrated machine learning and multimodal data fusion for patho-phenotypic feature recognition in iPSC models of dilated cardiomyopathy.

Integration of multiple data sources presents a challenge for accurate prediction of molecular patho-phenotypic features in automated analysis of data from human model systems. Here, we applied a machine learning-based data integration to distinguish patho-phenotypic features at the subcellular level for dilated cardiomyopathy (DCM). We employed a human induced pluripotent stem cell-derived cardiomyocyte (iPSC-CM) model of a DCM mutation in the sarcomere protein troponin T (TnT), TnT-R141W, compared to isogenic healthy (WT) control iPSC-CMs. We established a multimodal data fusion (MDF)-based analysis to integrate source datasets for Ca2+ transients, force measurements, and contractility recordings. Data were acquired for three additional layer types, single cells, cell monolayers, and 3D spheroid iPSC-CM models. For data analysis, numerical conversion as well as fusion of data from Ca2+ transients, force measurements, and contractility recordings, a non-negative blind deconvolution (NNBD)-based method was applied. Using an XGBoost algorithm, we found a high prediction accuracy for fused single cell, monolayer, and 3D spheroid iPSC-CM models (≥92±0.08 %), as well as for fused Ca2+ transient, beating force, and contractility models (>96±0.04 %). Integrating MDF and XGBoost provides a highly effective analysis tool for prediction of patho-phenotypic features in complex human disease models such as DCM iPSC-CMs.

GANtlitz: Ultra High Resolution Generative Model for Multi‐Modal Face Textures

AbstractHigh‐resolution texture maps are essential to render photoreal digital humans for visual effects or to generate data for machine learning. The acquisition of high resolution assets at scale is cumbersome, it involves enrolling a large number of human subjects, using expensive multi‐view camera setups, and significant manual artistic effort to align the textures. To alleviate these problems, we introduce GANtlitz (A play on the german noun Antlitz, meaning face), a generative model that can synthesize multi‐modal ultra‐high‐resolution face appearance maps for novel identities. Our method solves three distinct challenges: 1) unavailability of a very large data corpus generally required for training generative models, 2) memory and computational limitations of training a GAN at ultra‐high resolutions, and 3) consistency of appearance features such as skin color, pores and wrinkles in high‐resolution textures across different modalities. We introduce dual‐style blocks, an extension to the style blocks of the StyleGAN2 architecture, which improve multi‐modal synthesis. Our patch‐based architecture is trained only on image patches obtained from a small set of face textures (<100) and yet allows us to generate seamless appearance maps of novel identities at 6k × 4k resolution. Extensive qualitative and quantitative evaluations and baseline comparisons show the efficacy of our proposed system. (see https://www.acm.org/publications/class-2012)

Trends and Interannual Variability of the Hydroxyl Radical in the Remote Tropics During Boreal Autumn Inferred From Satellite Proxy Data

AbstractDespite its importance for the global oxidative capacity, spatially resolved trends and variability of the hydroxyl radical (OH) are poorly constrained. We demonstrate the utility of a tropospheric column OH (TCOH) product, created from machine learning and satellite proxy data, in determining the spatial variability in trends of tropical OH over the oceans during September through November. While OH increases domain‐wide by 2.1%/decade from 2005–2019, we find significant spatial heterogeneity in regional trends, with decreases in some areas of 2.5%/decade. Our analysis of the trends in the proxy data indicate anthropogenic‐driven changes in emissions of OH drivers as well as increasing temperatures cause these trends. This OH product is potentially a significant advance in constraining OH spatial variability and serves as a useful complement to existing tools in understanding the atmospheric oxidative capacity. Comprehensive observations of TCOH are required to assess the fidelity of this method.

AI-Powered Innovations in Electrical Engineering: Enhancing Efficiency, Reliability, and Sustainability

Artificial Intelligence (AI) has emerged as a transformative force in the field of Electrical Engineering, revolutionizing traditional practices and unlocking unprecedented possibilities. This research paper investigates the integration of AI-powered innovations to enhance efficiency, reliability, and sustainability within electrical engineering systems. Through a comprehensive review of existing literature, this study delves into key applications of AI, including predictive maintenance, optimal resource allocation, and fault detection, among others. Utilizing advanced machine learning algorithms and data analytics techniques, AI facilitates real-time decision-making processes, enabling proactive maintenance strategies and optimizing system performance. Moreover, AI-driven approaches contribute to the enhancement of reliability by predicting potential failures and implementing pre-emptive measures, consequently reducing downtime, and improving operational continuity. Furthermore, the implementation of AI in electrical engineering fosters sustainability by optimizing energy consumption, mitigating environmental impacts, and facilitating the integration of renewable energy sources into power grids. By leveraging AI technologies, electrical engineering systems can adapt to dynamic operational conditions, maximize resource utilization, and minimize environmental footprints, thereby paving the way for a more efficient, reliable, and sustainable future. This paper underscores the transformative potential of AI in shaping the landscape of electrical engineering and provides insights into future research directions to harness the benefits of AI-powered innovations further.

Machine Learning and Big Data Analytics for Precision Cardiac RiskStratification and Heart Diseases

The data explosion hasushered in a new era where insights are mined from vast data pools known as big data. Strategies for harnessing this data have emerged as critical decision-making tools across fields, employing various data analysis methods. Data mining techniques play an essential role in extracting meaningful patterns and insights. Thispaper focuses on the intersection of data mining and healthcare, particularly the critical concern of heart disease prediction.It presents a novel system that estimates heart attack risk, combining data mining with machine learning. Employing classification, the system stratifies data intotwo classes: heart disease presence or absence. Two powerful algorithms,decision tree classification and Naïve Bayes classification, enhance accuracy in predicting heart disease risk, achieving up to 91% and 87% accuracy, respectively. This review paper comprehensively analyzes the system's architecture, methodologies, and outcomes in healthcare, emphasizing data mining and machine learning's potential in medicine. Subsequent sections delve into methodology, results, and implications,providing a holistic view of this innovativeapproach.

Application of Machine Learning for Productivity Prediction in Tight Gas Reservoirs

Accurate well productivity prediction plays a significant role in formulating reservoir development plans. However, traditional well productivity prediction methods lack accuracy in tight gas reservoirs; therefore, this paper quantitatively evaluates the correlations between absolute open flow and the critical parameters for Linxing tight gas reservoirs through statistical analysis. Dominant control factors are obtained by considering reservoir engineering theories, and a novel machine learning-based well productivity prediction method is proposed for tight gas reservoirs. The adaptability of the productivity prediction model is assessed through machine learning and field data analysis. Combined with the typical decline curve analysis, the estimated ultimate recovery (EUR) of a single well in the tight gas reservoir is forecasted in an appropriate range. The results of the study include 10 parameters (such as gas saturation) identified as the dominant controlling factors for well productivity and geological factors that impact the productivity in this area compared to fracturing parameters. According to the prediction results of the three models, the R2 of Support Vector Regression (SVR), Back Propagation (BP), and Random Forest (RF) models are 0.72, 0.87, and 0.91, respectively. The results indicate that RF has a more accurate prediction. In addition, the RF model is more suitable for medium and high-production wells based on the actual field data. Based on this model, it is verified that the productivity of low-producing wells is affected by water production. This study confirms the model’s reliability and application value by predicting recoverable reserves for a single well.