Saved Queries

Using deep learning model predictions requires not only understanding the model’s confidence but also its uncertainty, so we know when to trust the prediction or require support from a human. In this study, we used Monte Carlo dropout (MCDO) to characterize the uncertainty of deep learning image classification algorithms, including feature fusion models, on simulated synthetic aperture radar (SAR) images of persistent ship wakes. Comparing to a baseline, we used the distribution of predictions from dropout with simple mean value ensembling and the Kolmogorov—Smirnov (KS) test to classify in-domain and out-of-domain (OOD) test samples, created by rotating images to angles not present in the training data. Our objective was to improve the classification robustness and identify OOD images during the test time. The mean value ensembling did not improve the performance over the baseline, in that there was a –1.05% difference in the Matthews correlation coefficient (MCC) from the baseline model averaged across all SAR bands. The KS test, by contrast, saw an improvement of +12.5% difference in MCC and was able to identify the majority of OOD samples. Leveraging the full distribution of predictions improved the classification robustness and allowed labeling test images as OOD. The feature fusion models, however, did not improve the performance over the single SAR-band models, demonstrating that it is best to rely on the highest quality data source available (in our case, C-band). Full article

►▼ Show Figures

Figure 1

11 pages, 408 KiB

Open AccessArticle

Domain Adversarial Convolutional Neural Network Improves the Accuracy and Generalizability of Wearable Sleep Assessment Technology

by Adonay S. Nunes, Matthew R. Patterson, Dawid Gerstel, Sheraz Khan, Christine C. Guo and Ali Neishabouri

Sensors 2024, 24(24), 7982; https://doi.org/10.3390/s24247982 (registering DOI) - 14 Dec 2024

Abstract

Wearable accelerometers are widely used as an ecologically valid and scalable solution for long-term at-home sleep monitoring in both clinical research and care. In this study, we applied a deep learning domain adversarial convolutional neural network (DACNN) model to this task and demonstrated that this new model outperformed existing sleep algorithms in classifying sleep–wake and estimating sleep outcomes based on wrist-worn accelerometry. This model generalized well to another dataset based on different wearable devices and activity counts, achieving an accuracy of 80.1% (sensitivity 84% and specificity 58%). Compared to commonly used sleep algorithms, this model resulted in the smallest error in wake after sleep onset (MAE of 48.7, Cole–Kripke of 86.2, Sadeh of 108.2, z-angle of 57.5) and sleep efficiency (MAE of 11.8, Cole–Kripke of 18.4, Sadeh of 23.3, z-angle of 9.3) outcomes. Despite being around for many years, accelerometer-alone devices continue to be useful due to their low cost, long battery life, and ease of use. Improving the accuracy and generalizability of sleep algorithms for accelerometer wrist devices is of utmost importance. We here demonstrated that domain adversarial convolutional neural networks can improve the overall accuracy, especially the specificity, of sleep–wake classification using wrist-worn accelerometer data, substantiating its use as a scalable and valid approach for sleep outcome assessment in real life. Full article

(This article belongs to the Section Wearables)

►▼ Show Figures

Figure 1

Figure 1
Model architecture. The DAsleepCNN model is composed of modules of a convolutional layer, batch normalization, and max pooling; after three modules the output is flattened and sent to a label classifier where it predicts sleep–wake labels for the MESA dataset, and to a domain adversarial classifier that classifies the dataset domain of the input. For MESA samples, the input labels are the dataset label and sleep–wake label, for NC samples only the dataset label is provided. For inference, the domain adversarial component is not used, only the label classifier. Full article ">Figure 2
Model accuracy. The accuracies of the four models presented are plotted for the NC and MESA datasets. For MESA, the highest accuracy was achieved by noDACNN25+25. However, this model had a marked drop in performance when applied on NC, showing poor generalizability. DACNN25+1, on the other hand, had a high accuracy on NC, which crucially was on par with its performance on MESA. Full article ">Figure 3
Confusion matrices and ROC-AUC. On top, the confusion matrix and ROC-AUC are plotted for the NC dataset using the best-performing model with input past 25 + 1. On the bottom, the same plots are shown for the MESA dataset with the model input of the past 25 + 25. Full article ">Figure 4
Average input values for correct and incorrect predictions in the datasets. The violin plots show the mean and 25th and 75th interquartile ranges for true and false predictions for the best-performing models. The left represents predictions from the NC dataset and the right from the MESA dataset. Full article ">

57 pages, 720 KiB

Open AccessReview

Exploring Kernel Machines and Support Vector Machines: Principles, Techniques, and Future Directions

by Ke-Lin Du, Bingchun Jiang, Jiabin Lu, Jingyu Hua and M. N. S. Swamy

Mathematics 2024, 12(24), 3935; https://doi.org/10.3390/math12243935 (registering DOI) - 13 Dec 2024

Abstract

The kernel method is a tool that converts data to a kernel space where operation can be performed. When converted to a high-dimensional feature space by using kernel functions, the data samples are more likely to be linearly separable. Traditional machine learning methods can be extended to the kernel space, such as the radial basis function (RBF) network. As a kernel-based method, support vector machine (SVM) is one of the most popular nonparametric classification methods, and is optimal in terms of computational learning theory. Based on statistical learning theory and the maximum margin principle, SVM attempts to determine an optimal hyperplane by addressing a quadratic programming (QP) problem. Using Vapnik–Chervonenkis dimension theory, SVM maximizes generalization performance by finding the widest classification margin within the feature space. In this paper, kernel machines and SVMs are systematically introduced. We first describe how to turn classical methods into kernel machines, and then give a literature review of exciting kernel machines.We then introduce the SVM model, its principles, and various SVM training methods for classification, clustering, and regression. Related topics, including optimizing model architecture, are also discussed. We conclude by outlining future directions for kernel machines and SVMs. This article functions both as a state-of-the-art survey and a tutorial. Full article

(This article belongs to the Special Issue Matrix Factorization for Signal Processing and Machine Learning)

16 pages, 1308 KiB

Open AccessArticle

Evaluating DL Model Scaling Trade-Offs During Inference via an Empirical Benchmark Analysis

by Demetris Trihinas, Panagiotis Michael and Moysis Symeonides

Future Internet 2024, 16(12), 468; https://doi.org/10.3390/fi16120468 (registering DOI) - 13 Dec 2024

Abstract

With generative Artificial Intelligence (AI) capturing public attention, the appetite of the technology sector for larger and more complex Deep Learning (DL) models is continuously growing. Traditionally, the focus in DL model development has been on scaling the neural network’s foundational structure to increase computational complexity and enhance the representational expressiveness of the model. However, with recent advancements in edge computing and 5G networks, DL models are now aggressively being deployed and utilized across the cloud–edge–IoT continuum for the realization of in situ intelligent IoT services. This paradigm shift introduces a growing need for AI practitioners, as a focus on inference costs, including latency, computational overhead, and energy efficiency, is long overdue. This work presents a benchmarking framework designed to assess DL model scaling across three key performance axes during model inference: classification accuracy, computational overhead, and latency. The framework’s utility is demonstrated through an empirical study involving various model structures and variants, as well as publicly available datasets for three popular DL use cases covering natural language understanding, object detection, and regression analysis. Full article

(This article belongs to the Special Issue Scalable and Distributed Cloud Continuum Orchestration for Next-Generation IoT Applications: Latest Advances and Prospects)

►▼ Show Figures

Figure 1

Figure 1
High-level overview of a deep neural network. Full article ">Figure 2
Pipeline of performance evaluation trade-offs. Full article ">Figure 3
Inference quality (classification accuracy and MSE) with respect to model complexity. The presented plots include: (a) BERT model variants, (b) EfficientNet model variants, and (c) MLP-Regression model variants. Full article ">Figure 4
Computational overhead of inference with respect to model complexity. The presented plots include: (a) BERT model variants, (b) EfficientNet model variants, and (c) MLP-Regression model variants. Full article ">Figure 5
Inference latency with respect to model complexity. The presented plots include: (a) BERT model variants, (b) EfficientNet model variants, and (c) MLP-Regression model variants. Full article ">

22 pages, 828 KiB

Open AccessArticle

MediScan: A Framework of U-Health and Prognostic AI Assessment on Medical Imaging

by Sibtain Syed, Rehan Ahmed, Arshad Iqbal, Naveed Ahmed and Mohammed Ali Alshara

J. Imaging 2024, 10(12), 322; https://doi.org/10.3390/jimaging10120322 - 13 Dec 2024

Abstract

With technological advancements, remarkable progress has been made with the convergence of health sciences and Artificial Intelligence (AI). Modern health systems are proposed to ease patient diagnostics. However, the challenge is to provide AI-based precautions to patients and doctors for more accurate risk assessment. The proposed healthcare system aims to integrate patients, doctors, laboratories, pharmacies, and administrative personnel use cases and their primary functions onto a single platform. The proposed framework can also process microscopic images, CT scans, X-rays, and MRI to classify malignancy and give doctors a set of AI precautions for patient risk assessment. The proposed framework incorporates various DCNN models for identifying different forms of tumors and fractures in the human body i.e., brain, bones, lungs, kidneys, and skin, and generating precautions with the help of the Fined-Tuned Large Language Model (LLM) i.e., Generative Pretrained Transformer 4 (GPT-4). With enough training data, DCNN can learn highly representative, data-driven, hierarchical image features. The GPT-4 model is selected for generating precautions due to its explanation, reasoning, memory, and accuracy on prior medical assessments and research studies. Classification models are evaluated by classification report (i.e., Recall, Precision, F1 Score, Support, Accuracy, and Macro and Weighted Average) and confusion matrix and have shown robust performance compared to the conventional schemes. Full article

(This article belongs to the Special Issue Deep Learning in Biomedical Image Segmentation and Classification: Advancements, Challenges and Applications)

25 pages, 844 KiB

Open AccessArticle

Enriching Value of Big Data Cooperative Assets from a Time-Horizon Perspective

by Shaobo Ren, Patrick S. W. Fong and Yi Zhang

Sustainability 2024, 16(24), 10961; https://doi.org/10.3390/su162410961 - 13 Dec 2024

Abstract

Driven by the rise of big data, enterprises urgently need to accurately utilize users’ real-time and accumulated information to realize present value and establish long-term advantages, then achieving the sustainable development. Previous works identified value co-created through big data as “big data cooperative assets”. However, while the mainstream research on this concept has primarily focused on analyzing its features, formation conditions, and influencing factors, particularly from the perspective of time-horizon value, an equally important area—the formation mechanism—has been neglected. To address this gap, this article constructs a classification framework of big data cooperative assets by combining time-horizon aspects with interaction dominators. It then examines the formation mechanisms of data link and data insight value through multi-case analysis. Overall, this research not only provides new perspectives for enriching the theoretical understanding of big data cooperative assets but also suggests useful practical guidelines for innovative interaction between enterprises and users in the age of data competition. In addition, improving the efficiency of realizing the value of big data cooperative assets helps the enterprise to better cope with external risks, such as market changes and policy adjustments, and maintain sound operations, further contributing to build a harmonious society and promote the construction of an ecological civilization. Full article

(This article belongs to the Special Issue Business and Collaborative Networks: Production, Creation and Shared Value)

►▼ Show Figures

Figure 1

Figure 1
The big data cooperative asset value classification framework. Full article ">Figure 2
Understanding value formation in big data cooperative assets. Full article ">

21 pages, 2608 KiB

Open AccessArticle

Voice Analysis in Dogs with Deep Learning: Development of a Fully Automatic Voice Analysis System for Bioacoustics Studies

by Mahmut Karaaslan, Bahaeddin Turkoglu, Ersin Kaya and Tunc Asuroglu

Sensors 2024, 24(24), 7978; https://doi.org/10.3390/s24247978 - 13 Dec 2024

Abstract

Extracting behavioral information from animal sounds has long been a focus of research in bioacoustics, as sound-derived data are crucial for understanding animal behavior and environmental interactions. Traditional methods, which involve manual review of extensive recordings, pose significant challenges. This study proposes an automated system for detecting and classifying animal vocalizations, enhancing efficiency in behavior analysis. The system uses a preprocessing step to segment relevant sound regions from audio recordings, followed by feature extraction using Short-Time Fourier Transform (STFT), Mel-frequency cepstral coefficients (MFCCs), and linear-frequency cepstral coefficients (LFCCs). These features are input into convolutional neural network (CNN) classifiers to evaluate performance. Experimental results demonstrate the effectiveness of different CNN models and feature extraction methods, with AlexNet, DenseNet, EfficientNet, ResNet50, and ResNet152 being evaluated. The system achieves high accuracy in classifying vocal behaviors, such as barking and howling in dogs, providing a robust tool for behavioral analysis. The study highlights the importance of automated systems in bioacoustics research and suggests future improvements using deep learning-based methods for enhanced classification performance. Full article

(This article belongs to the Special Issue Artificial Intelligence and Deep Learning in Sensors and Applications: 2nd Edition)

26 pages, 6618 KiB

Open AccessArticle

Monitoring Saltmarsh Restoration in the Upper Bay of Fundy Using Multi-Temporal Sentinel-2 Imagery and Random Forests Classifier

by Swarna M. Naojee, Armand LaRocque, Brigitte Leblon, Gregory S. Norris, Myriam A. Barbeau and Matthew Rowland

Remote Sens. 2024, 16(24), 4667; https://doi.org/10.3390/rs16244667 - 13 Dec 2024

Abstract

Saltmarshes provide important ecosystem services, including coastline protection, but face decline due to human activities and climate change. There are increasing efforts to conserve and restore saltmarshes worldwide. Our study evaluated the effectiveness of Sentinel-2 satellite imagery to monitor landcover changes using a saltmarsh restoration project undergoing its 9th to 12th year of recovery in the megatidal Bay of Fundy in Maritime Canada. Specifically, in 2019–2022, five satellite images per growing season were acquired. Random Forests classification for 13 landcover classes (ranging from bare mud to various plant communities) achieved a high overall classification accuracy, peaking at 96.43% in 2021. Field validation points confirmed this, with high validation accuracies reaching 93.02%. The classification results successfully distinguished ecologically significant classes, such as Spartina alterniflora–S. patens mix. Our results reveal the appearance of high marsh species in restoration sites and elevational-based zonation patterns, indicating progression. They demonstrate the potential of Sentinel-2 imagery for monitoring saltmarsh restoration projects in north temperate latitudes, aiding management efforts. Full article

►▼ Show Figures

Figure 1

Figure 1
Location of the 4 saltmarsh sites in Aulac, New Brunswick, in Sentinel-2 imagery acquired on 3 May 2021. A and D are the reference sites, and B and C the restoration sites. Full article ">Figure 2
Flowchart presenting the main image processing steps (input data in purple; image processing in light green, image classifier in pink; results in blue). Full article ">Figure 3
Landcover map of reference site A obtained by applying the RF classifier to multi-temporal Sentinel-2 images for 2019, 2020, 2021, and 2022. Full article ">Figure 3 Cont.
Landcover map of reference site A obtained by applying the RF classifier to multi-temporal Sentinel-2 images for 2019, 2020, 2021, and 2022. Full article ">Figure 4
Landcover map of reference site D obtained by applying the RF classifier to multi-temporal Sentinel-2 images for 2019, 2020, 2021, and 2022. Full article ">Figure 5
Landcover map of restoration site B obtained by applying the RF classifier to multi-temporal Sentinel-2 images for 2019, 2020, 2021, and 2022. Full article ">Figure 6
Landcover map of restoration site C obtained by applying the RF classifier to multi-temporal Sentinel-2 images for 2019, 2020, 2021, and 2022. Full article ">

39 pages, 925 KiB

Open AccessReview

Machine Learning Techniques for Sensor-Based Human Activity Recognition with Data Heterogeneity—A Review

by Xiaozhou Ye, Kouichi Sakurai, Nirmal-Kumar C. Nair and Kevin I-Kai Wang

Sensors 2024, 24(24), 7975; https://doi.org/10.3390/s24247975 - 13 Dec 2024

Abstract

Sensor-based Human Activity Recognition (HAR) is crucial in ubiquitous computing, analyzing behaviors through multi-dimensional observations. Despite research progress, HAR confronts challenges, particularly in data distribution assumptions. Most studies assume uniform data distributions across datasets, contrasting with the varied nature of practical sensor data in human activities. Addressing data heterogeneity issues can improve performance, reduce computational costs, and aid in developing personalized, adaptive models with fewer annotated data. This review investigates how machine learning addresses data heterogeneity in HAR by categorizing data heterogeneity types, applying corresponding suitable machine learning methods, summarizing available datasets, and discussing future challenges. Full article

(This article belongs to the Special Issue Non-Intrusive Sensors for Human Activity Detection and Recognition)

15 pages, 17108 KiB

Open AccessArticle

Investigations on the Performance of a 5 mm CdTe Timepix3 Detector for Compton Imaging Applications

by Juan S. Useche Parra, Gerardo Roque, Michael K. Schütz, Michael Fiederle and Simon Procz

Sensors 2024, 24(24), 7974; https://doi.org/10.3390/s24247974 - 13 Dec 2024

Abstract

Nuclear power plant decommissioning requires the rapid and accurate classification of radioactive waste in narrow spaces and under time constraints. Photon-counting detector technology offers an effective solution for the quick classification and detection of radioactive hotspots in a decommissioning environment. This paper characterizes a 5 CdTe Timepix3 detector and evaluates its feasibility as a single-layer Compton camera. The sensor’s electron mobility–lifetime product and resistivity are studied across bias voltages ranging from -100 to -3000, obtaining values of

μ_{e} τ_{e} = 1.2 (1) e - 3^{2} V^{- 1}

, and two linear regions with resistivities of

ρ_{I} = 5.8 (2) G Ω

and

ρ_{I I} = 4.1 (1) G Ω

. Additionally, two calibration methodologies are assessed to determine the most suitable for Compton applications, achieving an energy resolution of 16.3

k

e V

for the

^{137}

Cs photopeak. The electron’s drift time in the sensor is estimated to be 122.3 7.4

n

s

using cosmic muons. Finally, a Compton reconstruction of two simultaneous point-like sources is performed, demonstrating the detector’s capability to accurately locate radiation hotspots with a

\sim 51

resolution. Full article

(This article belongs to the Special Issue Recent Advances in X-Ray Sensing and Imaging)

17 pages, 222 KiB

Open AccessArticle

Practice and Prospect of Regulating Personal Data Protection in China

by Liping Yang, Yiling Lin and Bing Chen

Laws 2024, 13(6), 78; https://doi.org/10.3390/laws13060078 - 13 Dec 2024

Abstract

Privacy protection is a fundamental guarantee for secure data flows and the basic requirement for data security. A reasonable privacy protection system acts as a catalyst for unlocking the financial value of data. The current legislative framework for personal data protection in China, adhering to the principle of proportionality, establishes critical measures such as informed consent for data collection and processing, data classification and grading management, and remedies for data leakage and other risks. In addition, in judicial practice, typical disputes regarding personal information protection and privacy rights have been promoted to clarify the scope for collecting users’ personal information and biometric data. Although further improvements are needed in legislative, judicial, and technical approaches, China’s commitment and practice in personal data protection are noteworthy. The existing legislation, law enforcement, and technical practices play an increasingly vital role in realizing the financial value of data and are essential for international cooperation on privacy protection. Furthermore, it is crucial to actively explore cooperation mechanisms for cross-border data flows under the principle of data sovereignty, participate in developing international rules for cross-border data flows, and formulate different management norms for cross-border data flows across different industries. Full article

19 pages, 20601 KiB

Open AccessArticle

The Influence of Climate Change and Socioeconomic Transformations on Land Use and NDVI in Ordos, China

by Yin Cao, Zhigang Ye and Yuhai Bao

Atmosphere 2024, 15(12), 1489; https://doi.org/10.3390/atmos15121489 - 13 Dec 2024

Abstract

Land use change is related to a series of core issues of global environmental change, such as environmental quality improvement, sustainable utilization of resources, energy reuse and climate change. In this study, Google Earth Engine (GEE), a remote sensing natural environment monitoring and analysis platform, was used to realize the combination of Landsat TM/OLI data images with spectral features and topographic features, and the random forest machine learning classification method was used to supervise and classify the low-cloud composite image data of Ordos City. The results show that: (1) GEE has a powerful computing function, which can realize efficient and high-precision in-depth analysis of long-term multi-temporal remote sensing images and monitoring of land use change, and the accuracy of acquisition can reach 87%. Compared with other data sets in the same period, the overall and local classification results are more distinct than ESRI (Environmental Systems Research Institute) and GlobeLand 30 data products. Slightly lower than the Institute of Aerospace Information Innovation of the Chinese Academy of Sciences to obtain global 30 m of land cover fine classification products. (2) The overall accuracy of the land cover data of Ordos City from 2003 to 2023 is between 79–87%, and the Kappa coefficient is between 0.79–0.84. (3) Climate, terrain, population and other interactive factors combined with socio-economic population data and national and local policies are the main factors affecting land use change between 2003 and 2023. Full article

(This article belongs to the Special Issue Response of Vegetation to Climatic and Anthropogenic Drivers in the Plateau)

►▼ Show Figures

Figure 1

54 pages, 7881 KiB

Open AccessReview

Spectroscopy-Based Methods and Supervised Machine Learning Applications for Milk Chemical Analysis in Dairy Ruminants

by Aikaterini-Artemis Agiomavriti, Maria P. Nikolopoulou, Thomas Bartzanas, Nikos Chorianopoulos, Konstantinos Demestichas and Athanasios I. Gelasakis

Chemosensors 2024, 12(12), 263; https://doi.org/10.3390/chemosensors12120263 - 13 Dec 2024

Abstract

Milk analysis is critical to determine its intrinsic quality, as well as its nutritional and economic value. Currently, the advancements and utilization of spectroscopy-based techniques combined with machine learning algorithms have made the development of analytical tools and real-time monitoring and prediction systems in the dairy ruminant sector feasible. The objectives of the current review were (i) to describe the most widely applied spectroscopy-based and supervised machine learning methods utilized for the evaluation of milk components, origin, technological properties, adulterants, and drug residues, (ii) to present and compare the performance and adaptability of these methods and their most efficient combinations, providing insights into the strengths, weaknesses, opportunities, and challenges of the most promising ones regarding the capacity to be applied in milk quality monitoring systems both at the point-of-care and beyond, and (iii) to discuss their applicability and future perspectives for the integration of these methods in milk data analysis and decision support systems across the milk value-chain. Full article

(This article belongs to the Special Issue Artificial Intelligence (AI)/Machine Learning (ML)-Assisted Chemical Sensors)

►▼ Show Figures

Figure 1

22 pages, 10004 KiB

Open AccessArticle

High-Resolution Dynamic Monitoring of Rocky Desertification of Agricultural Land Based on Spatio-Temporal Fusion

by Xin Zhao, Zhongfa Zhou, Guijie Wu, Yangyang Long, Jiancheng Luo, Xingxin Huang, Jing Chen and Tianjun Wu

Land 2024, 13(12), 2173; https://doi.org/10.3390/land13122173 - 13 Dec 2024

Abstract

The current research on rocky desertification primarily prioritizes large-scale surveillance, with minimal attention given to internal agricultural areas. This study offers a comprehensive framework for bedrock extraction in agricultural areas, employing spatial constraints and spatio-temporal fusion methodologies. Utilizing the high resolution and capabilities of Gaofen-2 imagery, we first delineate agricultural land, use these boundaries as spatial constraints to compute the agricultural land bedrock response Index (ABRI), and apply the spatial and temporal adaptive reflectance fusion model (STARFM) to achieve spatio-temporal fusion of Gaofen-2 imagery and Sentinel-2 imagery from multiple time periods, resulting in a high-spatio-temporal-resolution bedrock discrimination index (ABRI*) for analysis. This work demonstrates the pronounced rocky desertification phenomenon in the agricultural land in the study area. The ABRI* effectively captures this phenomenon, with the classification accuracy for the bedrock, based on the ABRI* derived from Gaofen-2 imagery, reaching 0.86. The bedrock exposure area in the farmland showed a decreasing trend from 2019 to 2021, a significant increase from 2021 to 2022, and a gradual decline from 2022 to 2024. Cultivation activities have a significant impact on rocky desertification within agricultural land. The ABRI significantly enhances the capabilities for the dynamic monitoring of rocky desertification in agricultural areas, providing data support for the management of specialized farmland. For vulnerable areas, timely adjustments to planting schemes and the prioritization of intervention measures such as soil conservation, vegetation restoration, and water resource management could help to improve the resilience and stability of agriculture, particularly in karst regions. Full article

►▼ Show Figures

Figure 1

16 pages, 361 KiB

Open AccessArticle

Stroke Dataset Modeling: Comparative Study of Machine Learning Classification Methods

by Kalina Kitova, Ivan Ivanov and Vincent Hooper

Algorithms 2024, 17(12), 571; https://doi.org/10.3390/a17120571 - 13 Dec 2024

Abstract

Stroke prediction is a vital research area due to its significant implications for public health. This comparative study offers a detailed evaluation of algorithmic methodologies and outcomes from three recent prominent studies on stroke prediction. Ivanov et al. tackled issues of imbalanced datasets and algorithmic bias using deep learning techniques, achieving notable results with a 98% accuracy and a 97% recall rate. They utilized resampling methods to balance the classes and advanced imputation techniques to handle missing data, underscoring the critical role of data preprocessing in enhancing the performance of Support Vector Machines (SVMs). Hassan et al. addressed missing data and class imbalance using multiple imputations and the Synthetic Minority Oversampling Technique (SMOTE). They developed a Dense Stacking Ensemble (DSE) model with over 96% accuracy. Their results underscore the efficiency of ensemble learning techniques and imputation for handling imbalanced datasets in stroke prediction. Bathla et al. employed various classifiers and feature selection techniques, including SMOTE, for class balancing. Their Random Forest (RF) classifier, combined with Feature Importance (FI) selection, achieved an accuracy of 97.17%, illustrating the positive impact of RF and relevant feature selection on model performance. A comparative analysis indicated that Ivanov et al.’s method achieved the highest accuracy rate. However, the studies collectively highlight that the choice of models and techniques for stroke prediction should be tailored to the specific characteristics of the dataset used. This study emphasizes the importance of effective data management and model selection in enhancing predictive performance. Full article

(This article belongs to the Special Issue Algorithms in Data Classification (2nd Edition))

►▼ Show Figures

Figure 1

Figure 1
Sparsity matrix for the dataset. Full article ">

Show export options Show export options

Select all

Export citation of selected articles as:

Error

Oops... you haven't selected anything for export.

Displaying article 1-50 on page 1 of 1007.

Go to page 1 2 3 4 5

Search Results (50,324)

Further Information

Guidelines

MDPI Initiatives

Follow MDPI