Search Results (7,583)

The power quality index is an important index in the power industry. Power quality disturbances (PQDs) have a great impact on the power grid. It is important to identify the type of PQDs accurately. However, it is difficult to analyze a large number of PQDs, especially in more complex power systems. Considering the limitations of the traditional time–frequency domain method and the complexity of the optimization algorithm in extracting features, a novel algorithm is proposed to classify the PQDs in this paper, which is based on the unscented Kalman filter (UKF) and the kernel extreme learning machine (KELM). The UKF is used to detect and process the original disturbances, and anti-noise detection performance is analyzed by tracking the amplitude change in voltage swell under different signal–noise ratios (SNRs). The amplitudes of the fundamental wave, third harmonic, fifth harmonic, seventh harmonic, and oscillatory wave are tracked in real time, and their minimum and peak indexes are taken as the optimal feature vector set. The feature vector set is classified by the KELM. The performance of the proposed method has also been evaluated with simulated and experimental results. Full article

We developed a method for the evaluation and selection of customer business analysis in two stages. First, using the bank’s existing expert model, artificial rules of thumb were used to evaluate the value of each field of the data and establish screening rules. Secondly, the machine learning feature screening method was applied based on the customer’s transaction data to find out whether the customer’s contribution to the bank had a significant impact as a feature of the model. Based on the results, the best classification model was selected through data verification. The effectiveness of the proposed model was validated through actual case analysis, taking wealth management in banks as an example. The classification method, using support vector machines (SVMs), effectively assists banks in identifying potential customers efficiently and in planning to manage customers. This method helps to avoid the traditional blind spots, which emerge based on subjective judgment, and allows bank wealth managers to promote customer relationship management (CRM). Full article

To achieve intelligent manufacturing and improve the machining quality of machine tools, this paper proposes an interpretable machining size prediction model combining eXtreme Gradient Boosting (XGBoost), autoencoder (AE), and Shapley additive explanation (SHAP) analysis. In this study, XGBoost was used to establish an evaluation system for the actual machining size of computer numerical control (CNC) machine tools. The XGBoost model was combined with SHAP approximation to effectively capture local and global features in the data using autoencoders and transform the preprocessed data into more representative feature vectors. Grey correlation analysis (GRA) and principal component analysis (PCA) were used to reduce the dimensions of the original data features, and the synthetic minority overstimulation technique of the Gaussian noise regression (SMOGN) method was used to deal with the problem of data imbalance. Taking the actual size of the machine tool as the response parameter, based on the size parameters in the milling process of the CNC machine tool, the effectiveness of the model is verified. The experimental results show that the proposed AE-XGBoost model is superior to the traditional XGBoost method, and the prediction accuracy of the model is 7.11% higher than that of the traditional method. The subsequent SHAP analysis reveals the importance and interrelationship of features and provides a reliable decision support system for machine tool processing personnel, helping to improve processing quality and achieve intelligent manufacturing. Full article

In view of the problem that the recommendation system is not good at integrating multi-source information and user sentiment, this paper proposes a BERT-LSTM Dual-Tower Recommendation Method for Sequential Feature Extraction (BLDRM-SFE). This method uses BERT to extract semantic features from user reviews and item details and obtains vector representations of item IDs and their groups through embedding. The user tower combines user review features with item group features to generate a user vector, while the item tower integrates item detail features with item ID vectors to generate an item vector. The outputs of the two towers are processed by LSTM to handle item ID sequence information, uncover potential sequence relationships, and generate rating predictions, thereby constructing a personalized recommendation list. The experimental results show that this method significantly outperforms four baseline models—BERT4Rec, PRM, BST, and ComiRec—on the Amazon Review Data and Yelp datasets. On the Amazon dataset, BLDRM-SFE improves by 10.39%, 8.08%, 10.78%, 10.59%, and 5.49% across five metrics compared to the baseline models; on the Yelp dataset, the improvements are 10.95%, 10.06%, 13.04%, 12.59%, and 10.8%, respectively. In addition, ablation experiments confirmed the positive impact of item ID sequence information on the method’s performance. The results show that the incorporation of sequence information significantly enhanced the recommendation performance. Full article

Background/Objectives: Pathway annotations of non-macromolecular (relatively small) biomolecules facilitate biological and biomedical interpretation of metabolomics datasets. However, low pathway annotation levels of detected biomolecules hinder this type of interpretation. Thus, predicting the pathway involvement of detected but unannotated biomolecules has a high potential to improve metabolomics data analysis and omics integration. Past publications have only made use of the Kyoto Encyclopedia of Genes and Genomes-derived datasets to develop machine learning models to predict pathway involvement. However, to our knowledge, the Reactome knowledgebase has not been utilized to develop these types of predictive models. Methods: We created a dataset ready for machine learning using chemical representations of all pathway-annotated compounds available from the Reactome knowledgebase. Next, we trained and evaluated a multilayer perceptron binary classifier using combined metabolite-pathway paired feature vectors engineered from this new dataset. Results: While models trained on a prior corresponding KEGG dataset with 502 pathways scored a mean Matthew’s correlation coefficient (MCC) of 0.847 and a 0.0098 standard deviation, the models trained on the Reactome dataset with 3985 pathways demonstrated improved performance with a mean MCC of 0.916, but with a higher standard deviation of 0.0149. Conclusions: These results indicate that the pathways in Reactome can also be effectively predicted, greatly increasing the number of human-defined pathways available for prediction. Full article

Recently, extensive research efforts have concentrated on comprehending the semantic features of images in the field of computer vision. In order to address the spatial orientation relations among water surface targets (WSTs) in an image, which is a fundamental semantic feature, this paper focused on the recognition of spatial orientation relations. We first developed the water surface target spatial orientation vector field (WST-SOVF) algorithm, a novel end-to-end methodology, to recognize these spatial orientation relations among WSTs in an image. The WST-SOVF algorithm encodes the spatial orientation relation into the learning framework of a new deep convolutional neural network model, which comprises two distinct branches: the T-branch and the S-branch, both designed for the spatial feature extraction. The T-branch employs keypoint estimation to identify central points and classify the WST categories, while the S-branch constructs a spatial orientation vector field between WSTs, where each pixel in the field encodes the spatial orientation angle between two separated WSTs and collectively determines the category of spatial orientation. A fusion module was also designed to integrate the spatial feature obtained from both branches, thereby generating a comprehensive triple list that provides not only all the WSTs and their spatial orientation relations, but also their associated confidence levels. We performed a comparative evaluation of our WST-SOVF algorithm based on Huawei’s “Typical Surface/Underwater Target Recognition” dataset and the results demonstrated the outstanding performance of WST-SOVF algorithm. Full article

Charactering alteration and its geochemical signature provides critical information relevant to ore-deposit genesis and its related footprint; for porphyry-type deposits, zoned potassic-phyllic-propylitic alteration and metal enrichment are critical features. Here we integrate earlier lithological and mineralogical studies of the (10+ Moz Au) Archean Côté Gold porphyry-type Au(-Cu) deposit (Ontario, Canada) with identified alteration types to provide exploration vectors. The ca. 2740 tonalite-quartz diorite-diorite intrusive complex and co-temporal Au(-Cu) mineralization as disseminations, breccias and veins are co-spatial with ore-related alteration types (amphibole, biotite, muscovite). An early, locally developed amphibole event coring the deposit is followed by emplacement of a Au(-Cu) mineralized biotite-rich magmatic-hydrothermal breccia body and broad halo of disseminated biotite and quartz veining. These rocks record gains via mass balance calculations of K, Fe, Mg, LILE, and LREE with Au, Cu, Mo, Ag, Se and Bi. Later muscovite alteration is enriched in K, Rb, Cs, Ba, CO₂, and LOI with varied Au, Cu, Mo, Te, As, and Bi values. A strong albite overprint records extreme Na gains with the loss of most other elements, including ore metals (i.e., Au, Cu). Together these data define an Au-Cu-Mo-Ag-Te-Bi-Se core co-spatial with biotite breccia versus a peripheral stockwork and sheeted vein zone with a Te-Se-Zn-Pb-As association. These features further support the posited porphyry-type model for the Côté Gold Au(-Cu) deposit. Full article

This study analyzes collapse hazards for complex interactions between geology, meteorology, and human activities in the Changbai Mountain region, focusing on how to cope with these features through machine learning. Using a dataset of 651 collapse events, this study evaluates four machine learning methods, Support Vector Machine (SVM), Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine (LightGBM), to deal with complex nonlinear data structures. To overcome the limitations of a single-feature selection method, a variance inflation factor is introduced to optimize the selection of collapse risk factors. The transparency and interpretability of the modeling results are enhanced by combining the Shapley Additive Explanations (SHAP) with interpretable artificial intelligence. Model performance is evaluated on a test set by several statistical metrics, which shows that the optimized random forest model performs best and outperforms SVM, XGBoost, and LightGBM. The SHAP analysis results indicate that distance from the road is a key factor for collapse hazard. This study emphasizes the need for collapse management strategies that provide interpretable solutions for collapse hazard assessment. Full article

This study explores the application of advanced machine learning (ML) models to predict CO₂ solubility in NaCl brine, a critical parameter for effective carbon capture, utilization, and storage (CCUS). Using a comprehensive database of 1404 experimental data points spanning temperature (−10 to 450 °C), pressure (0.098 to 140 MPa), and salinity (0.017 to 6.5 mol/kg), the research evaluates the predictive capabilities of five ML algorithms: Decision Tree, Random Forest, XGBoost, Multilayer Perceptron, and Support Vector Regression with a radial basis function kernel. Among these, XGBoost demonstrated the highest overall accuracy, achieving an R² value of 0.9926, with low root mean square error (RMSE) and mean absolute error (MAE) of 0.0655 and 0.0191, respectively. A feature importance analysis revealed that pressure has the most impactful effect and positively correlates with CO₂ solubility, while temperature generally exhibits a negative effect. A higher accuracy was found when the developed model was compared with one well-established empirical model and one ML-based model from the literature. The results underscore the potential of ML models to significantly enhance prediction accuracy over a wide data range, reduce computational costs, and improve the efficiency of CCUS operations. This work demonstrates the robustness and adaptability of ML approaches for modeling complex subsurface conditions, paving the way for optimized carbon sequestration strategies. Full article

With the widespread use of credit cards in online and offline transactions, credit card fraud has become a significant challenge in the financial sector. The rapid advancement of payment technologies has led to increasingly sophisticated fraud techniques, necessitating more effective detection methods. While machine learning has been extensively applied in fraud detection, the application of deep learning methods remains relatively limited. Inspired by brain-like computing, this work employs the Continuous-Coupled Neural Network (CCNN) for credit card fraud detection. Unlike traditional neural networks, the CCNN enhances the representation of complex temporal and spatial patterns through continuous neuron activation and dynamic coupling mechanisms. Using the Kaggle Credit Card Fraud Detection (CCFD) dataset, we mitigate data imbalance via the Synthetic Minority Oversampling Technique (SMOTE) and transform sample feature vectors into matrices for training. Experimental results show that our method achieves an accuracy of 0.9998, precision of 0.9996, recall of 1.0000, and an F1-score of 0.9998, surpassing traditional machine learning models, which highlight CCNN’s potential to enhance the security and efficiency of fraud detection in the financial industry. Full article

Objective: The objective of this study was to early-detect gross motor abnormalities through video detection in Taiwanese infants aged 2–6 months. Background: The current diagnosis of infant developmental delays primarily relies on clinical examinations. However, during clinical visits, infants may show atypical behaviors due to unfamiliar environments, which might not truly reflect their true developmental status. Methods: This study utilized videos of infants recorded in their home environments. Two pediatric neurologists manually annotated these clips to identify whether an infant possessed the characteristics of gross motor delays through an assessment of his/her gross motor movements. Using transfer learning techniques, four pose recognition models, including ViTPose, HRNet, DARK, and UDP, were applied to the infant gross motor dataset. Four machine learning classification models, including random forest, support vector machine, logistic regression, and XGBoost, were used to predict the developmental status of infants. Results: The experimental results of pose estimation and tracking indicate that the ViTPose model provided the best performance for pose recognition. A total of 227 features related to kinematics, motions, and postures were extracted and calculated. A one-way ANOVA analysis revealed 106 significant features that were retained for constructing prediction models. The results show that a random forest model achieved the best performance with an average F1-score of 0.94, a weighted average AUC of 0.98, and an average accuracy of 94%. Full article

Multivariate Time Series Classification (MTSC) is a challenging task in real-world applications. Current approaches emphasize modeling multiscale relationships over time. However, the Multivariate Time Series (MTS) also exhibits multiscale cross-channel relationships. Furthermore, the long-term temporal relationships in time series are difficult to capture. In this paper, we introduce MD-Former, a Multiscale Dual-Branch Attention network leveraging the Transformer architecture to capture multiscale relationships across time and channels for MTSC. In MD-Former, MTS is embedded into 2D vectors using Channel-Patching (CP) to retain channel information. Following this, we develop two branches: the Interlaced Attention Branch (IAB) and the Channel-Independent Attention Branch (CIAB). The IAB facilitates the fusion of information across channels and time, while the CIAB prevents the loss of information resulting from excessive fusion. Both the IAB and CIAB consist of multiple layers, each representing a distinct time scale. Finally, we utilize features from each layer of both IAB and CIAB as inputs to the Multiscale Classification Head (MCH) for feature fusion and classification. Experimental results show that MD-Former achieves performance levels that are comparable to SOTA methods in MTSC. Full article

Mass spectrometry (MS) data present challenges for machine learning (ML) classification due to their high dimensionality, complex feature distributions, batch effects, and intensity discrepancies, often hindering model generalization and efficiency. To address these issues, this study introduces the Efficient Quick 1D Lite Convolutional Neural Network (CNN) Ensemble Classifier (EQLC-EC), integrating 1D convolutional networks with reshape layers and dual voting mechanisms for enhanced feature representation and classification performance. Validation was performed on five publicly available MS datasets, each featured in high-impact publications. EQLC-EC underwent comprehensive evaluation against classical machine learning (ML) models (e.g., support vector machine (SVM), random forest) and the leading deep learning methods reported in these studies. EQLC-EC demonstrated dataset-specific improvements, including enhanced classification accuracy (1–5% increase) and reduced standard deviation (1–10% reduction). Performance differences between soft and hard voting mechanisms were negligible (<1% variation in accuracy and standard deviation). EQLC-EC presents a powerful and efficient tool for MS data analysis with potential applications across metabolomics and proteomics. Full article

Buried hill reservoirs are characterized by complex formation conditions and highly heterogeneous rock structures, which result in the poor performance of traditional crossplot methods in stratigraphic lithology classification. Logging curves, as comprehensive reflections of various petrophysical properties, are influenced by complex geological factors, leading to overlapping response values even among different lithologies with similar physical properties. This overlap negatively impacts the accuracy of intelligent lithology identification methods. To address this challenge, this study leverages logging response data, experimental data, and mud logging data to propose an optimized inversion method for mineral content, introducing mineral curves to resolve the curve overlap issue. By analyzing six wells in the study area, models were constructed using the calculated mineral content curves and conventional logging features to mitigate the feature overlap. The XGBoost algorithm was employed to identify lithologies by addressing the nonlinear relationships inherent in complex reservoirs. The experimental results indicate that the optimized mineral curves significantly enhance the model’s discriminative capability, effectively addressing the decline in identification accuracy due to feature overlap. Compared to models such as Random Forest (RF) and Support Vector Machine (SVM), the XGBoost model demonstrated superior accuracy and stability, providing a reliable basis for precise reservoir identification in the study area. Full article

Effective pig farming relies on precise and adaptable animal identification methods, particularly in dynamic environments where new pigs are regularly added to the herd. However, pig face recognition is challenging due to high individual similarity, lighting variations, and occlusions. These factors hinder accurate identification and monitoring. To address these issues under Open-Set conditions, we propose a three-phase Pig Face Open-Set Recognition (PFOSR) system. In the Training Phase, we adopt a decoupled design, first training a YOLOv8-based pig face detection model on a small labeled dataset to automatically locate pig faces in raw images. We then refine a Vision Transformer (ViT) recognition model via a dual-loss strategy—combining Sub-center ArcFace and Center Loss—to enhance both inter-class separation and intra-class compactness. Next, in the Known Pig Registration Phase, we utilize the trained detection and recognition modules to extract representative embeddings from 56 identified pigs, storing these feature vectors in a Pig Face Feature Gallery. Finally, in the Unknown and Known Pig Recognition and Registration Phase, newly acquired pig images are processed through the same detection–recognition pipeline, and the resulting embeddings are compared against the gallery via cosine similarity. If the system classifies a pig as unknown, it dynamically assigns a new ID and updates the gallery without disrupting existing entries. Our system demonstrates strong Open-Set recognition, achieving an AUROC of 0.922, OSCR of 0.90, and F1-Open of 0.94. In the closed set, it attains a precision@1 of 0.97, NMI of 0.92, and mean average precision@R of 0.96. These results validate our approach as a scalable, efficient solution for managing dynamic farm environments with high recognition accuracy, even under challenging conditions. Full article