Kernel Support Research Articles

Machine learning complete intersection Calabi-Yau 3-folds

Gaussian process regression, kernel support vector regression, the random forest, extreme gradient boosting, and the generalized linear model algorithms are applied to data of complete intersection Calabi-Yau threefolds. It is shown that Gaussian process regression is the most suitable for learning the Hodge number h2,1 in terms of h1,1. The performance of this regression algorithm is such that the Pearson correlation coefficient for the validation set is R2=0.9999999995 with a root mean square error RMSE=0.0002895011. As for the train set, these two parameters are as follows: R2=0.9999999994 and RMSE=0.0002854348. The training error and the cross-validation error of this regression are 1×10−9 and 1.28×10−7, respectively. Learning the Hodge number h1,1 in terms of h2,1 yields R2=1.000000 and RMSE=7.395731×10−5 for the validation set of the Gaussian process regression. Published by the American Physical Society 2024

Prediction of Post Traumatic Epilepsy Using MR-Based Imaging Markers.

Post-traumatic epilepsy (PTE) is a debilitating neurological disorder that develops after traumatic brain injury (TBI). Despite the high prevalence of PTE, current methods for predicting its occurrence remain limited. In this study, we aimed to identify imaging-based markers for the prediction of PTE using machine learning. Specifically, we examined three imaging features: Lesion volumes, resting-state fMRI-based measures of functional connectivity, and amplitude of low-frequency fluctuation (ALFF). We employed three machine-learning methods, namely, kernel support vector machine (KSVM), random forest, and an artificial neural network (NN), to develop predictive models. Our results showed that the KSVM classifier, with all three feature types as input, achieved the best prediction accuracy of 0.78 AUC (area under the receiver operating characteristic (ROC) curve) using nested cross-validation. Furthermore, we performed voxel-wise and lobe-wise group difference analyses to investigate the specific brain regions and features that the model found to be most helpful in distinguishing PTE from non-PTE populations. Our statistical analysis uncovered significant differences in bilateral temporal lobes and cerebellum between PTE and non-PTE groups. Overall, our findings demonstrate the complementary prognostic value of MR-based markers in PTE prediction and provide new insights into the underlying structural and functional alterations associated with PTE.

Authenticity markers of cultivated blueberries (Vaccinium sp.) and European wild blueberry (V. myrtillus) assessed by hyperspectral imaging and machine learning algorithms

The present study was undertaken to tackle food fraud in the food chain by developing a novel panel of analytical fingerprints to distinguish the fruit of bilberry (Vaccinium myrtillus), a wild European blueberry grown in Finland, from their commercial cultivated counterparts based on V. corymbosum or other originally American Vaccinium species. The analytical fingerprints were derived using hyperspectral imaging (HSI). The cultivated domestic blueberry samples from Finland (N = 188), those imported (N = 338), and domestic Finnish wild bilberry fruit (N = 247) were imaged using Specim IQ and FX17 HSI cameras. The fruit HSI data were then classified using seven machine learning model types. The supervised evolutionary feature selection algorithm was used to recognize the most separating subset of wavelengths (nm predictors, colors) between the fruit class labels, which were 1) wild bilberry fruit vs. cultivated blueberry fruit and 2) domestic cultivated blueberry fruit vs. imported cultivated blueberry fruit. Moreover, we ran 190,822 individual models and 54,137 predictor combinations to determine whether subsets or all wavelengths better distinguished fruit class labels. Overall, the best model type, the rbf kernel support vector machine (rbf-SVM), showed excellent prediction performance when classifying wild bilberries vs. cultivated blueberries with HSI data obtained using both camera types (4.1% and 3.2% holdout set errors). Also the rbf-SVM performed well when classifying domestic cultivated blueberries vs. those imported with HSI data obtained using an IQ camera (2.3% holdout set error). Our results showed that hyperspectral imaging could predict food fraud attempts concerning the Vaccinium fruit of different origin and species.

Ionospheric gradient anomaly detection using kernel support vector machine for ground-based augmentation systems

Ionospheric gradient anomaly detection using kernel support vector machine for ground-based augmentation systems

Architecture of an automated program complex based on a multiple kernel svm classifier for analyzing malicious executable files

Subject matter. This article presents the development and architecture of an automated program complex designed to identify and analyze malicious executable files using a classifier based on a multiple kernel support vector machine (SVM). Goal. The aim of the work is to create an automated system that enhances the accuracy and efficiency of malware detection by combining static and dynamic analysis into a single framework capable of processing large volumes of data with optimal time expenditure. Tasks. To achieve this goal, tasks were carried out that included developing a program complex that automates the collection of static and dynamic data from executable files using tools like IDA Pro, IDAPython, and Drakvuf; integrating a multiple kernel SVM classifier to analyze the collected heterogeneous data; validating the system's effectiveness based on a substantial dataset containing 1,389 executable samples; and demonstrating the system's scalability and practical applicability in real-world conditions. Methods. The methods involved a hybrid approach that combines static analysis – extracting byte code, disassembled instructions, and control flow graphs using IDA Pro and IDAPython – with dynamic analysis, which entails monitoring real-time behavior using Drakvuf. The multiple kernel SVM classifier integrates different data representations using various kernels, allowing for both linear and nonlinear relationships to be considered in the classification process. Results. The results of the study show that the system achieves a high level of accuracy and completeness, as evidenced by key performance metrics such as an F-score of 0.93 and ROC AUC and PR AUC values. The automated program complex reduces the analysis time of a single file from an average of 11 minutes to approximately 5 minutes, effectively doubling the throughput compared to previous methods. This significant reduction in processing time is critically important for deployment in environments where rapid and accurate malware detection is necessary. Furthermore, the system's scalability allows for efficient processing of large data volumes, making it suitable for real-world applications. Conclusions. In conclusion, the automated program complex developed in this study demonstrates significant improvements in the accuracy and efficiency of malware detection. By integrating multiple kernel SVM classification with static and dynamic analysis, the system shows potential for real-time malware detection and analysis. Its scalability and practical applicability indicate that it could become an important tool in combating modern cyber threats, providing organizations with an effective means to enhance their cybersecurity.

Critical physics-informed fatigue life prediction of laser 3D printed AlSi10Mg alloys with mass internal defects

The significant scatter in high cycle fatigue life of additively manufactured metallic components presents an increasing challenge to structural integrity. This fatigue life variation is radically attributed to the differences in physical features of critical defects that lead to crack initiation. To address this issue, this paper proposes an integrated framework for identifying critical defects and predicting fatigue life using physics-informed machine learning, with a focus on the impact of 3D defect features. By employing X-ray tomography, high cycle fatigue tests, and fractography analyses on post-mortem specimens, a dataset associated with mass internal defects is first built up to correlate the spatial geometric features of critical defects with fatigue life. A kernel support vector machine is then used to formulate a critical defect identification model, aimed at identifying critical defects among numerous defects by evaluating their geometric attributes. Finally, a fatigue life prediction model is developed using a physics-informed neural network, which incorporates the influence of defect geometry on fatigue life as physical constraints in the loss function. The integrated framework demonstrates that fatigue life predictions from identified critical defects in each specimen exhibit small deviations, with the average prediction falling within twice the error bands. This study is expected to provide a valuable reference for fatigue assessment of additively manufactured components through sequential critical defect identification and fatigue life prediction.

Co-training machine learning enables interpretable discovery of near-infrared phosphors with high performance

Near-infrared (NIR) phosphors based on Cr3+ doped garnets present great potential in the next generation of NIR light sources. Nevertheless, the huge searching space for the garnet composition makes the rapid discovery of NIR phosphors with high performance remain a great challenge for the scientific community. Herein, a generalizable machine learning (ML) strategy is designed to accelerate the exploration of innovative NIR phosphors via establishing the relationship between key parameters and emission peak wavelength (EPW). We propose a semi-supervised co-training model based on kernel ridge regression (KRR) and support vector regression (SVR), which successfully establishes an expanded dataset with unlabeled dataset (previously unidentified garnets), addressing the overfitting issue resulted from a small dataset and greatly improving the model generalization capability. The model is then interpreted to extract valuable insights into the contribution originated from different features. And a new type NIR luminescent material of Lu3Y2Ga3O12: Cr3+ (EPW~750 nm) is efficiently screened, which demonstrates a high internal (external) quantum efficiency of 97.1% (38.8%) and good thermal stability, particularly exhibiting promising application in the NIR phosphor-converted LEDs (pc-LED). These results suggest the strategy proposed in this work could provide new viewpoint and direction for developing NIR luminescence materials.

Maximum Local Density-Driven Non-overlapping Radial Basis Function Support Kernel Neural Network

The learning and optimization of kernels in the radial basis function neural network (RBFNN) are crucial. However, in existing methods, there are issues of overfitting when learning kernel parameters. The learned kernels are also sensitive to outliers. This paper proposes a general kernel learning strategy for RBFNN called non-overlapping maximum local density support kernel learning (MLD-SKL), which contains two modules, the non-overlapping maximum local density (MLD) kernel learning module and support kernel learning (SKL) module. In the MLD kernel learning stage, the candidate set of kernels is incrementally determined based on the local density of samples. Meanwhile, it is required that the coverage ranges of kernels from different classes do not overlap with each other. This module is effective in reducing the impact of outliers. In the SKL stage, kernel importance indicator is defined to measure the importance of kernels. The learned support kernels are utilized to construct a maximum local density-driven non-overlapping radial basis function support kernel neural network (MLD-RBFSKNN). The RBFNN constructed through MLD-SKL exhibits a more compact structure. The experiments demonstrate that the proposed MLD-RBFSKNN improves accuracy in recognition task. Furthermore, while achieving superior recognition performance, the final constructed network also has the minimum number of kernels.

Lithium Battery Thermal-Runaway Monitoring Based on Whole-Feature Neural Networks

Thermal runaway monitoring and analysis has become a serious challenge to the safety of lithium-ion battery driven electric equipment. Thermal-runaway monitoring is crucial to avoid the burning and explosion of lithium batteries. This paper proposes a new type of deep neural network, known as whole-feature neural networks (WFNN), for lithium battery thermal-runaway monitoring. The neural networks learn the thermal-runaway patterns of a lithium battery from the measured temperatures, current, and voltages. WFNN is an end-to-end model for thermal-runaway monitoring of lithium batteries. An experiment on thermal-runaway monitoring of lithium batteries was carried out to evaluate the performance of the proposed WFNN. The monitoring accuracy is up to 99.48%, which is higher than those of support vector machine, kernel support vector machine, k-nearest neighbor, and fully-connected neural networks. Moreover, the computation efficiency of WFNN is high enough for real-time thermal-runaway monitoring. As a result, experimental results show that the proposed WFNN is applicable to the thermal-runaway monitoring of lithium batteries.

A Novel Dual Kernel Support Vector-Based Levy Dung Beetle Algorithm for Accurate Speech Emotion Detection

A Novel Dual Kernel Support Vector-Based Levy Dung Beetle Algorithm for Accurate Speech Emotion Detection

Classification of LAMOST spectra of B-type and hot subdwarf stars using kernel support vector machine

Machine learning has emerged as a leading field in artificial intelligence, demonstrating expert-level performance in various domains. Astronomy has benefited from machine learning techniques, particularly in classifying and identifying stars based on their features. This study focuses on the spectra-based classification of 11,408 B-type and 2422 hot subdwarf stars. The study employs baseline correction using Asymmetric Least Squares (ALS) to enhance classification accuracy. It applies the Pan-Core concept to identify 500 unique patterns or ranges for both types of stars. These patterns are the foundation for creating Support Vector Machine (SVM) models, including the linear (L-SVM), polynomial (P-SVM), and radial basis (R-SVM) kernels. Parameter tuning for the SVM models is achieved through cross-validation. Evaluation of the SVM models on test data reveals that the linear kernel SVM achieves the highest accuracy (87.0%), surpassing the polynomial kernel SVM (84.1%) and radial kernel SVM (80.1%). The average calibrated accuracy falls within the range of 90–95%. These results demonstrate the potential of using spectrum-based classification to aid astronomers in improving and expanding their understanding of stars, with a specific focus on the identification of hot subdwarf stars. This study presents a valuable investigation for astronomers, as it enables the classification of stars based on their spectra, leveraging machine learning techniques to enhance their knowledge and insights in astronomy.

Leveraging Quantum Kernel Support Vector Machine for breast cancer diagnosis from Digital Breast Tomosynthesis images

Leveraging Quantum Kernel Support Vector Machine for breast cancer diagnosis from Digital Breast Tomosynthesis images

Electroencephalographic Classification Reveals Atypical Speech Motor Planning in Stuttering Adults.

This study explores speech motor planning in adults who stutter (AWS) and adults who do not stutter (ANS) by applying machine learning algorithms to electroencephalographic (EEG) signals. In this study, we developed a technique to holistically examine neural activity differences in speaking and silent reading conditions across the entire cortical surface. This approach allows us to test the hypothesis that AWS will exhibit lower separability of the speech motor planning condition. We used the silent reading condition as a control condition to isolate speech motor planning activity. We classified EEG signals from AWS and ANS individuals into speaking and silent reading categories using kernel support vector machines. We used relative complexities of the learned classifiers to compare speech motor planning discernibility for both classes. AWS group classifiers require a more complex decision boundary to separate speech motor planning and silent reading classes. These findings indicate that the EEG signals associated with speech motor planning are less discernible in AWS, which may result from altered neuronal dynamics in AWS. Our results support the hypothesis that AWS exhibit lower inherent separability of the silent reading and speech motor planning conditions. Further investigation may identify and compare the features leveraged for speech motor classification in AWS and ANS. These observations may have clinical value for developing novel speech therapies or assistive devices for AWS.

A novel temporal-frequency combination pattern optimization approach based on information fusion for motor imagery BCIs

Motor imagery (MI) stands as a powerful paradigm within Brain-Computer Interface (BCI) research due to its ability to induce changes in brain rhythms detectable through common spatial patterns (CSP). However, the raw feature sets captured often contain redundant and invalid information, potentially hindering CSP performance. Methodology-wise, we propose the Information Fusion for Optimizing Temporal-Frequency Combination Pattern (IFTFCP) algorithm to enhance raw feature optimization. Initially, preprocessed data undergoes simultaneous processing in both time and frequency domains via sliding overlapping time windows and filter banks. Subsequently, we introduce the Pearson-Fisher combinational method along with Discriminant Correlation Analysis (DCA) for joint feature selection and fusion. These steps aim to refine raw electroencephalogram (EEG) features. For precise classification of binary MI problems, an Radial Basis Function (RBF)-kernel Support Vector Machine classifier is trained. To validate the efficacy of IFTFCP and evaluate it against other techniques, we conducted experimental investigations using two EEG datasets. Results indicate a notably superior classification performance, boasting an average accuracy of 78.14% and 85.98% on dataset 1 and dataset 2, which is better than other methods outlined in this article. The study’s findings suggest potential benefits for the advancement of MI-based BCI strategies, particularly in the domain of feature fusion.

Combination of static and dynamic neural imaging features to distinguish sensorineural hearing loss: a machine learning study.

Sensorineural hearing loss (SNHL) is the most common form of sensory deprivation and is often unrecognized by patients, inducing not only auditory but also nonauditory symptoms. Data-driven classifier modeling with the combination of neural static and dynamic imaging features could be effectively used to classify SNHL individuals and healthy controls (HCs). We conducted hearing evaluation, neurological scale tests and resting-state MRI on 110 SNHL patients and 106 HCs. A total of 1,267 static and dynamic imaging characteristics were extracted from MRI data, and three methods of feature selection were computed, including the Spearman rank correlation test, least absolute shrinkage and selection operator (LASSO) and t test as well as LASSO. Linear, polynomial, radial basis functional kernel (RBF) and sigmoid support vector machine (SVM) models were chosen as the classifiers with fivefold cross-validation. The receiver operating characteristic curve, area under the curve (AUC), sensitivity, specificity and accuracy were calculated for each model. SNHL subjects had higher hearing thresholds in each frequency, as well as worse performance in cognitive and emotional evaluations, than HCs. After comparison, the selected brain regions using LASSO based on static and dynamic features were consistent with the between-group analysis, including auditory and nonauditory areas. The subsequent AUCs of the four SVM models (linear, polynomial, RBF and sigmoid) were as follows: 0.8075, 0.7340, 0.8462 and 0.8562. The RBF and sigmoid SVM had relatively higher accuracy, sensitivity and specificity. Our research raised attention to static and dynamic alterations underlying hearing deprivation. Machine learning-based models may provide several useful biomarkers for the classification and diagnosis of SNHL.

Evaluating Explainable Machine Learning Models for Clinicians

Gaining clinicians’ trust will unleash the full potential of artificial intelligence (AI) in medicine, and explaining AI decisions is seen as the way to build trustworthy systems. However, explainable artificial intelligence (XAI) methods in medicine often lack a proper evaluation. In this paper, we present our evaluation methodology for XAI methods using forward simulatability. We define the Forward Simulatability Score (FSS) and analyze its limitations in the context of clinical predictors. Then, we applied FSS to our XAI approach defined over an ML-RO, a machine learning clinical predictor based on random optimization over a multiple kernel support vector machine (SVM) algorithm. To Compare FSS values before and after the explanation phase, we test our evaluation methodology for XAI methods on three clinical datasets, namely breast cancer, VTE, and migraine. The ML-RO system is a good model on which to test our XAI evaluation strategy based on the FSS. Indeed, ML-RO outperforms two other base models—a decision tree (DT) and a plain SVM—in the three datasets and gives the possibility of defining different XAI models: TOPK, MIGF, and F4G. The FSS evaluation score suggests that the explanation method F4G for the ML-RO is the most effective in two datasets out of the three tested, and it shows the limits of the learned model for one dataset. Our study aims to introduce a standard practice for evaluating XAI methods in medicine. By establishing a rigorous evaluation framework, we seek to provide healthcare professionals with reliable tools for assessing the performance of XAI methods to enhance the adoption of AI systems in clinical practice.

Two-step machine learning-assisted label-free surface-enhanced Raman spectroscopy for reliable prediction of dissolved furfural in transformer oil

Accurate detection of dissolved furfural in transformer oil is crucial for real-time monitoring of the aging state of transformer oil-paper insulation. While label-free surface-enhanced Raman spectroscopy (SERS) has demonstrated high sensitivity for dissolved furfural in transformer oil, challenges persist due to poor substrate consistency and low quantitative reliability. Herein, machine learning (ML) algorithms were employed in both substrate fabrication and spectral analysis of label-free SERS. Initially, a high-consistency Ag@Au substrate was prepared through a combination of experiments, particle swarm optimization-neural network (PSO-NN), and a hybrid strategy of particle swarm optimization and genetic algorithm (Hybrid PSO-GA). Notably, a two-step ML framework was proposed, whose operational mechanism is classification followed by quantification. The framework adopts a hierarchical modeling strategy, incorporating simple algorithms such as kernel support vector machine (Kernel-SVM), k-nearest neighbors (KNN), etc., to independently establish lightweight regression models on each cluster, which allows each model to focus more effectively on fitting the data within its cluster. The classification model achieved an accuracy of 100%, while the regression models exhibited an average correlation coefficient (R2) of 0.9953 and the root mean square errors (RMSE) consistently below 10-2. Thus, this ML framework emerges as a rapid and reliable method for detecting dissolved furfural in transformer oil, even in the presence of different interfering substances, which may also have potentiality for other complex mixture monitoring systems.

A novel machine learning model for efficacy prediction of immunotherapy-chemotherapy in NSCLC based on CT radiomics

Lung cancer is categorized into two main types: non-small cell lung cancer (NSCLC) and small cell lung cancer. Of these, NSCLC accounts for approximately 85% of all cases and encompasses varieties such as squamous cell carcinoma and adenocarcinoma. For patients with advanced NSCLC that do not have oncogene addiction, the preferred treatment approach is a combination of immunotherapy and chemotherapy. However, the progression-free survival (PFS) typically ranges only from about 6 to 8 months, accompanied by certain adverse events. In order to carry out individualized treatment more effectively, it is urgent to accurately screen patients with PFS for more than 12 months under this treatment regimen. Therefore, this study undertook a retrospective collection of pulmonary CT images from 60 patients diagnosed with NSCLC treated at the First Affiliated Hospital of Wenzhou Medical University. It developed a machine learning model, designated as bSGSRIME-SVM, which integrates the rime optimization algorithm with self-adaptive Gaussian kernel probability search (SGSRIME) and support vector machine (SVM) classifier. Specifically, the model initiates its process by employing the SGSRIME algorithm to identify pivotal image features. Subsequently, it utilizes an SVM classifier to assess these features, aiming to enhance the model's predictive accuracy. Initially, the superior optimization capability and robustness of SGSRIME in IEEE CEC 2017 benchmark functions were validated. Subsequently, employing color moments and gray-level co-occurrence matrix methods, image features were extracted from images of 60 NSCLC patients undergoing immunotherapy combined with chemotherapy. The developed model was then utilized for analysis. The results indicate a significant advantage of the model in predicting the efficacy of immunotherapy combined with chemotherapy for NSCLC, with an accuracy of 92.381% and a specificity of 96.667%. This lays the foundation for more accurate PFS predictions and personalized treatment plans.

Machine learning approaches to debris flow susceptibility analyses in the Yunnan section of the Nujiang River Basin.

The Yunnan section of the Nujiang River (YNR) Basin in the alpine-valley area is one of the most critical areas of debris flow in China. We analyzed the applicability of three machine learning algorithms to model of susceptibility to debris flow-Random Forest (RF), the linear kernel support vector machine (Linear SVM), and the radial basis function support vector machine (RBFSVM)-and compared 20 factors to determine the dominant controlling in debris flow occurrence in the region. We found that (1) RF outperformed RBFSVM and Linear SVM in terms of accuracy, (2) topographic conditions were prerequisites, and geology, precipitation, vegetation, and anthropogenic influence were critical to forming debris flows. Also, the relative elevation difference was the most prominent evaluation factor of debris flow susceptibility, and (3) susceptibility maps based on RF's debris flow susceptibility (DFS) showed that zones with very high susceptibility were distributed along the mainstream of the Nujiang River. These findings provide methodological guidance and reference for improvement of DFS assessment. It enriches the content of DFS studies in the alpine-valley areas.

Machine Learning Insights: Exploring Key Factors Influencing Sale-to-List Ratio—Insights from SVM Classification and Recursive Feature Selection in the US Real Estate Market

The US real estate market is a complex ecosystem influenced by multiple factors, making it critical for stakeholders to understand its dynamics. This study uses Zillow Econ (monthly) data from January 2018 to October 2023 across 100 major regions gathered through Metropolitan Statistical Area (MSA) and advanced machine learning techniques, including radial kernel Support Vector Machines (SVMs), used to predict the sale-to-list ratio, a key metric that indicates the market health and competitiveness of the US real estate. Recursive Feature Elimination (RFE) is used to identify influential variables that provide insight into market dynamics. Results show that SVM achieves approximately 85% accuracy, with temporal indicators such as Days to Pending and Days to Close, pricing dynamics such as Listing Price Cut and Share of Listings with Price Cut, and rental market conditions captured by the Zillow Observed Rent Index (ZORI) emerging as critical factors influencing the sale-to-list ratio. The comparison between SVM alphas and RFE highlights the importance of time, price, and rental market indicators in understanding market trends. This study underscores the interplay between these variables and provides actionable insights for stakeholders. By contextualizing the findings within the existing literature, this study emphasizes the importance of considering multiple factors in housing market analysis. Recommendations include using pricing dynamics and rental market conditions to inform pricing strategies and negotiation tactics. This study adds to the body of knowledge in real estate research and provides a foundation for informed decision-making in the ever-evolving real estate landscape.