Learning Methods Research Articles

Cutting-Edge strategies for absence data identification in natural hazards: Leveraging Voronoi-Entropy in flood susceptibility mapping with advanced AI techniques

One of the non-structural methods for flood management is preparing flood susceptibility mapping (FSM). The performance of flood susceptibility models significantly depends on the learning methods and data sampling. Random sampling is commonly employed for data sampling owing to its practicality and straightforwardness. A purely random method may not be the best choice for non-flood sampling, as floods typically occur in different locations at different times. Therefore, this research aims to propose a new method for determining flood absence points using a combination of Voronoi diagrams and the entropy method. To achieve this, flood susceptibility modeling was conducted using the XGBoost (eXtreme Gradient Boosting) algorithm optimized with the cat swarm optimization (CSO) algorithm, employing the proposed method for absence point determination in the sub-basin of the Khuzestan province of Iran. Therefore, for flood susceptibility modeling, three scenarios for determining no-occurrence points were employed: random sampling (Scenario 1), Voronoi diagram (Scenario 2), and the combination of Voronoi diagram with entropy-based method (Scenario 3). Additionally, three data split ratios (60:40, 70:30, and 80:20) were utilized for partitioning training and validation datasets.Scenario 1 demonstrates varying area under the curve (AUC) of the receiver operating characteristic (ROC) curve values across different data split ratios, with the 60:40 ratio showing moderate accuracy (AUC = 0.720). Scenario 2 exhibits improved performance with higher AUC values (0.847 to 0.908) and balanced sensitivity–specificity trade-offs. Scenario 3 demonstrates varying performance across different data split ratios, achieving the highest modeling accuracy with an AUC of 0.888 in the 60:40 split. Overall, Scenarios 2 and 3 outperform Scenario 1, showcasing significant accuracy improvements ranging from 23.33 % to 30.1 % across different data split ratios. Using Voronoi diagrams and entropy-based methods notably enhances accuracy in determining no-occurrence points compared to random selection, emphasizing the importance of method selection in flood susceptibility modeling.

Implementing Mental Practice in Postgraduate Surgical Training for Minimally Invasive Surgery: A Systematic Review and Thematic Analysis.

Unprecedented pressure on the National Health Service (NHS) has meant that there are increasing obstacles to surgical training. Simulation training is an option to improve surgical performance but is limited due to availability, accessibility and financial constraints. Mental practice (MP) has been proposed as a potential solution to supplement the traditional method of apprenticeship-style learning. Despite increasing evidence that MP may be a useful tool to improve surgical performance and reduce surgeon anxiety, it is not widely adopted. This systematic review and thematic analysis aims to identify key themes that would allow for the successful implementation of MP in postgraduate surgical training. Medline, Embase and PsycINFO databases were systematically searched to identify articles that investigate the role of MP in improving surgical performance amongst surgical trainees. Retrieved papers were studied to inform thematic analysis of their content and studies were assessed for bias. A total of 321 studies were retrieved, of which 11 met the inclusion criteria. Overall risk of bias was assessed to be between critical and moderate for seven nonrandomized studies and between fair and good for four randomized studies. Key themes were identified and discussed using a thematic analysis approach. This study has identified that attaining high quality mental imagery is fundamental to success in mental practice and this can be augmented by the use of relaxation therapy and/or motivational imagery. Future research should focus on the application of MP in real-world surgical practice and breaking down complex procedures into fewer operative steps.

Machine learning-based sales forecasting during crises: Evidence from a Turkish women's clothing retailer.

Retail involves directly delivering goods and services to end consumers. Natural disasters and epidemics/pandemics have significant potential to disrupt supply chains, leading to shortages, forecasting errors, price increases, and substantial financial strains on retailers. The COVID-19 pandemic highlighted the need for retail sectors to prepare for crisis impacts on sales forecasts by regularly assessing and adjusting sales volumes, consumer behavior, and forecasting models to adapt to changing conditions. This study explores strategies for adapting sales forecasts and retail approaches in response to such crises. By employing different machine learning (ML) methods, we analyze consumer behavior changes and sales impacts across various product categories, including bottom wear, top wear, one piece, accessories, outwear, and shoes during the COVID-19 pandemic. The gradient boosting and CatBoost algorithms excelled in product groups with significant sales changes during the pandemic. The Multi-Layer Perceptron (MLP) algorithm performed well in low-volume categories like accessories and footwear. Meanwhile, MLP, LightGBM, and XGBoost were effective in medium-volume categories such as outerwear and underwear. The findings highlight the efficacy of these models in adapting sales forecasts to crisis conditions, offering a practical approach to enhancing retail resilience against future disruptions. This study offers an effective approach for adapting sales forecasting to shifting consumer behaviors during crises.

Outlier-resistant guided wave dispersion curve recovery and measurement placement optimization base on multitask complex hierarchical sparse Bayesian learning

Due to extensive detection range and high sensitivity to defects, ultrasonic Lamb waves are extensively studied in the fields of Nondestructive Testing and Structural Health Monitoring. In scenarios where the material parameters or geometric parameters of the waveguide are unknown, the dispersion relation of the guided wave cannot be calculated by the forward model. Consequently, it becomes imperative to extract wave propagation characteristics of Lamb wave from the acquired Lamb wave data. This paper presents a multitask complex hierarchical sparse Bayesian learning (MuCHSBL) method which is aimed at enhancing the efficacy of the dispersion relation solution by considering the continuity of the recovered dispersion curve in the frequency-wavenumber domain. Furthermore, the posterior distributions quantified by MuCHSBL are employed to optimize the placement of measurement points. Numerical and experimental studies are conducted to verify the effectiveness of the proposed method. Comparison analysis with the conventional approach demonstrates the significant enhancement in accuracy of recovering dispersion curves by the proposed method.

Chemotherapy selection for advanced or metastatic pancreatic cancer using machine learning models trained with multi-center datasets.

738 Background: Several factors must be considered when selecting the chemotherapeutic regimen for advanced or metastatic pancreatic cancer. The decision between FOLFIRINOX and Gemcitabine/Nab-paclitaxel (GnP) as the first-line therapy is challenging, as survival is influenced by the efficacy and toxicity profiles of these treatments, along with individual patient characteristics and therapeutic vulnerabilities. This study aims to facilitate the selection of an appropriate first-line regimen for advanced or metastatic pancreatic cancer by employing machine learning (ML) methods trained on multi-center datasets. Methods: Our study is based on a cohort of 191 patients who underwent systemic chemotherapy for advanced or metastatic pancreatic cancer at the Gangneung Asan Hospital and Asan Medical Center in South Korea between 2014 and 2023. The dataset consisted of 17 types of demographic and clinical characteristics, as well as time-course of response and survival outcomes. We divided the cohort into training and test groups (4:1) in a stratified manner. ML models predicting overall survival (OS) were trained on the former group using the CatBoost method. The models utilizing a minimal subset of variables were selected with respect to ROC-AUC during 5-fold cross validation. A patient was classified as having high risk to a therapy when the predicted probability of death following treatment initiation was above the threshold value as determined during training of the respective model. Results: The median age of the entire cohort was 62 years, with 64% being male. The ML models achieved ROC-AUCs of 0.81 and 0.82 when predicting early death within 12 months following the initial administration of FOLFIRINOX or GnP, respectively. The predictive accuracies of the models were 0.77 and 0.80 for the unseen datasets from the two treatment groups, respectively. The median OS of the high- versus low-risk groups of FOLFIRINOX predicted by the ML models were significantly different (9 vs 15 months, P <0.0001), with a hazard ratio (HR) of 2.8. Similarly, the median OS of the high- and low-risk groups of GnP were significantly different (9 vs 18 months, P <0.01, HR of 2.5). We observed that 27% of the patients predicted as high risk to FOLFIRINOX therapy were predicted as low risk to GnP and vice versa for 32% of the patients predicted as high risk to GnP therapy. Conclusions: We developed ML models to compute the probability of early death following FOLFIRINOX or GnP therapy from routinely collected data of the patients with advanced or metastatic pancreatic cancer. After confirming robust predictive performance with multi-center datasets, we believe that these models may assist clinicians in selecting a first-line regimens, potentially enhancing personalized treatment strategies for this challenging disease.

Machine learning models for quantitatively prediction of toxicity in macrophages induced by metal oxide nanoparticles.

As nanotechnology advances, metal oxide nanoparticles (MeONPs) increasingly come into contact with humans. The inhaled MeONPs cannot be effectively cleared by cilia or lung mucus. In the last decade, potential immune toxicity arising from exposure to MeONPs has been extensively debated, as lung macrophage is the main pathway for cleaning inhaled exogenous particles. However, their toxicity on lung macrophages has rarely been quantitatively predicted in silico due to the complexity of responses in macrophages and the intricate properties of MeONPs. Here, machine learning (ML) methods were used to establish models for quantitatively predicting the toxicity of MeONPs in macrophages. A multidimensional dataset including 240 data points covering the lethality, biochemical behaviors, and physicochemical properties of 30 MeONPs was obtained. ML models based on different algorithms with high prediction accuracy were constructed by addressing the issue of class imbalance during the training process. The models were verified by 10-fold cross-validation and external validation. The best-performed model has an R2 of 0.85 and 0.90 in the 10-fold cross-validation and external test set, respectively; and Q2 of 0.88 and 0.90 in the 10-fold cross-validation and test set, respectively. Five parameters that impact toxicity were identified and the toxicity mechanisms were elucidated by ML analysis. The prediction results can be used to fill the data gap in the risk assessment of nanomaterials. The framework offers valuable insights for designing and utilizing safe nanoparticles, as well as aiding in decision-making processes aimed at protecting the environment and public health.

Handling class imbalance in education using data-level and deep learning methods

In the current field of education, universities must be highly competitive to thrive and grow. Education data mining has helped universities in bringing in new students and retaining old ones. However, there is a major issue in this task, which is the class imbalance between the successful students and at-risk students that causes inaccurate predictions. To address this issue, 12 methods from data-level sampling techniques and 2 methods from deep learning synthesizers were compared against each other and an ideal class balancing method for the dataset was identified. The evaluation was done using the light gradient boosting machine ensemble model, and the metrics included receiver operating characteristic curve, precision, recall and F1 score. The two best methods were Tomek links and neighbourhood cleaning rule from undersampling technique with a F1 score of 0.72 and 0.71 respectively. The results of this paper identified the best class balancing method between the two approaches and identified the limitations of the deep learning approach.

Implementing cloud computing in drug discovery and telemedicine for quantitative structure-activity relationship analysis

This work aims to use cutting-edge machine learning methods to improve quantitative structure-activity relationship (QSAR) analysis, which is used in drug development and telemedicine. The major goal is to examine the performance of several predictive modeling approaches, including random forest, deep learning-based QSAR models, and support vector machines (SVM). It investigates the potential of feature selection techniques developed in chemoinformatics for enhancing model accuracy. The innovative aspect is using cloud computing resources to strengthen computational skills, allowing for managing massive amounts of chemical information. This strategy produces accurate and generalizable QSAR models. By using the cloud's scalability and constant availability, remote healthcare apps have a workable answer. The goal is to show how these methods may improve telemedicine and the drug development process. Utilizing cloud computing equips researchers with a flexible set of tools for precise and timely QSAR analysis, speeding up the discovery of bioactive chemicals for therapeutic use. This new method fits well with the dynamic nature of pharmaceutical study and has the potential to transform the way drugs are developed and delivered to patients via telemedicine.

Semi-supervised single-view 3D reconstruction via multi shape prior fusion strategy and self-attention

In the domain of single-view 3D reconstruction, traditional techniques have frequently relied on expensive and time-intensive 3D annotation data. Facing the challenge of annotation acquisition, semi-supervised learning strategies offer an innovative approach to reduce the dependence on labeled data. Despite these developments, the utilization of this learning paradigm in 3D reconstruction tasks remains relatively constrained. In this research, we created an innovative semi-supervised framework for 3D reconstruction that distinctively uniquely introduces a multi shape prior fusion strategy, intending to guide the creation of more realistic object structures. Additionally, to improve the quality of shape generation, we integrated a self-attention module into the traditional decoder. In benchmark tests on the ShapeNet dataset, our method substantially outperformed existing supervised learning methods at diverse labeled ratios of 1%, 10%, and 20%. Moreover, it showcased excellent performance on the real-world Pix3D dataset. Through comprehensive experiments on ShapeNet, our framework demonstrated a 3.3% performance improvement over the baseline. Moreover, stringent ablation studies further confirmed the notable effectiveness of our approach. Our code has been released on https://github.com/NWUzhouwei/SSMP.

Toward 3D hedonic price model for vertically developed cities using street view images and machine learning methods

Toward 3D hedonic price model for vertically developed cities using street view images and machine learning methods

Machine learning methods for compression capacity prediction and sensitivity analysis of concrete-filled steel tubular columns: State-of-the-art review

Machine learning methods for compression capacity prediction and sensitivity analysis of concrete-filled steel tubular columns: State-of-the-art review

FABLDroid: Malware detection based on hybrid analysis with factor analysis and broad learning methods for android applications

FABLDroid: Malware detection based on hybrid analysis with factor analysis and broad learning methods for android applications

Integrating direct and indirect views for group recommendation: An inter- and intra-view contrastive learning method

Integrating direct and indirect views for group recommendation: An inter- and intra-view contrastive learning method

Deep Learning techniques to detect and analysis of multiple sclerosis through MRI: A systematic literature review.

Deep learning (DL) techniques represent a rapidly advancing field within artificial intelligence, gaining significant prominence in the detection and analysis of various medical conditions through the analysis of medical data. This study presents a systematic literature review (SLR) focused on deep learning methods for the detection and analysis of multiple sclerosis (MS) using magnetic resonance imaging (MRI). The initial search identified 401 articles, which were rigorously screened, a selection of 82 highly relevant studies. These selected studies primarily concentrate on key areas such as multiple sclerosis, deep learning, convolutional neural networks (CNN), lesion segmentation, and classification, reflecting their alignment with the current state of the art. This review comprehensively examines diverse deep-learning approaches for MS detection and analysis, offering a valuable resource for researchers. Additionally, it presents key insights by summarizing these DL techniques for MS detection and analysis using MRI in a structured tabular format.

Enhancing meteorological data reliability: An explainable deep learning method for anomaly detection.

Accurate meteorological observation data is of great importance to human production activities. Meteorological observation systems have been advancing toward automation, intelligence, and informatization. Yet, instrumental malfunctions and unstable sensor node resources could cause significant deviations of data from the actual characteristics it should reflect. To achieve greater data accuracy, early detections of data anomalies, continuous collections and timely transmissions of data are essential. While obvious anomalies can be readily identified, the detection of systematic and gradually emerging anomalies requires further analyses. This study develops an interpretable deep learning method based on an autoencoder (AE), SHapley Additive exPlanations (SHAP) and Bayesian optimization (BO), in order to facilitate prompt and accurate anomaly detections of meteorological observational data. The proposed method can be unfolded into four parts. Firstly, the AE performs anomaly detections based on multidimensional meteorological datasets by marking the data that shows significant reconstruction errors. Secondly, the model evaluates the importance of each meteorological element of the flagged data via SHapley Additive exPlanation (SHAP). Thirdly, a K-sigma based threshold automatic delineation method is employed to obtain reasonable anomaly thresholds that are subject to the data characteristics of different observation sites. Finally, the BO algorithm is adopted to fine-tune difficult hyperparameters, enhancing the model's structure and thus the accuracy of anomaly detection. The practical implication of the proposed model is to inform agricultural production, climate observation, and disaster prevention.

MMVSL: A multi-modal visual semantic learning method for pig pose and action recognition

MMVSL: A multi-modal visual semantic learning method for pig pose and action recognition

Machining parameter optimization for a batch milling system using multi-task deep reinforcement learning

The integrated multi-objective optimization of machining parameters for improved machining quality and efficiency is important in batch milling systems. Due to the change of the batch milling system state, the continuous use of the same machining parameters may lead to degradation in quality and efficiency for workpieces in batches. Machining parameter optimization is usually determined by manual experience or trial-and-error methods, making it difficult to achieve a synergistic consideration of both quality and efficiency. To address this issue, a novel multi-task deep reinforcement learning method for machining parameter optimization in a batch machining system is proposed. Firstly, a reliable parallel joint estimation model of multiple machining quality and efficiency indicators is established using a multi-task time series estimation method, which can learn the correlation of these indicators to improve estimation accuracy. Then, the parameter optimization problem is formalized as a Markov decision process supported by a reinforcement learning virtual environment and an agent. The reinforcement learning virtual environment with the joint estimation model is constructed to improve the accuracy of optimized machining parameters for the collaborative optimization of quality and efficiency indicators. Within the virtual environment, time series sequential state, sequential action, multi-objective reward function, and constraint conditions adapted to the joint estimation model are defined to repeatedly evaluate different machining parameters. The agent with a multi-head attention and a dynamic weight adjustment mechanism is designed to improve the stability of the optimization process. Finally, experiments on a real machining dataset of thin-walled parts show that compared with the traditional deep reinforcement learning algorithm, the optimization effect of the proposed framework is improved by 9%-12%, and the standard deviation is decreased by 9% -18%.

A versatile feature selection learning method for satellite attitude control system fault diagnosis with limited data

A versatile feature selection learning method for satellite attitude control system fault diagnosis with limited data

A real-time prediction framework for energy consumption of electric buses using integrated Machine learning algorithms

An accurate prediction of energy consumption in electric buses (EBs) can effectively reduce driving range anxiety and facilitate bus scheduling. Existing studies have not provided real-time predictions based on distance traveled using integrated machine learning methods. This study proposes a framework for predicting EB energy consumption, which is primarily divided into energy consumption estimation, kinematic feature prediction, and energy consumption prediction. The framework begins by fusing high-resolution real-world EB data with weather and road information, from which five types of influencing factors are extracted for different driving distances. An eXtreme Gradient Boosting (XGBoost) model is developed to evaluate feature importance and estimate the energy consumption rate (ECR). The SHapley Additive explanation (SHAP) method is then used to analyze the factors affecting the ECR. To predict important kinematic characteristics, spatial and temporal characteristics are captured using Long Short-Term Memory (LSTM) and a fully connected neural network. Finally, the predicted kinematic characteristics and the XGBoost model are combined to enable real-time prediction of the ECR. The results indicate that estimation and prediction accuracies gradually improve with increased driving distance. The mean absolute error of average ECR decreases from 43.9 % for 100 m to 7.5 % for 16 km. Temperature, bus stop density, and peak periods emerge as the most significant external factors after 8 km. This framework shows an improvement of over 10 % in most scenarios compared with other models in the literature, enabling individual forecasts of energy consumption currently in transit and aiding in the calculation of remaining battery-supported distance.

A comparative analysis of the health monitoring process using deep learning methods for brain tumour

A comparative analysis of the health monitoring process using deep learning methods for brain tumour