Bayesian Hyperparameter Optimization Research Articles

An Investigation into the Susceptibility to Landslides Using Integrated Learning and Bayesian Optimization: A Case Study of Xichang City

In the middle southern section of the Freshwater River–Small River Fault system, Xichang City, Daliang Prefecture, Sichuan Province, is situated in the junction between the Anning River Fault and the Zemu River Fault. There has been a risk of increased activity in the fault zone in recent years, and landslide susceptibility evaluation for the area can effectively reduce the risk of disaster occurrence. Using integrated learning and Bayesian hyperparameter optimization, 265 landslides in Xichang City were used as samples in this study. Thirteen influencing factors were chosen to assess landslide susceptibility, and the BO-XGBoost, BO-LightGBM, and BO-RF models were evaluated using precision, recall, F1, accuracy, and AUC curves. The findings indicated that after removing the terrain relief evaluation factor, the four most significant factors associated with landslide susceptibility were NDVI, distance from faults, slope, and distance from rivers. The study demonstrates that the AUC value of the BO-XGBoost model in the study area is 0.8677, demonstrating a better generalization ability and higher prediction accuracy than the BO-LightGBM and BO-RF models. After Bayesian optimization of hyperparameters, the model offers a significant improvement in prediction accuracy.

Prospective silent deployment and evaluation of an intelligent machine learning model for prediction of emergency department visits during cancer treatment.

407 Background: Patients undergoing cancer treatment often need to visit the emergency department (ED), straining the healthcare system. We aim to use long-term electronic health record (EHR) data to deploy and evaluate a previously built warning system designed to accurately identify those at risk of ED visits. This system, once validated, will enable clinicians to take early, personalized actions to prevent these ED visits, saving resources and enhancing the quality of life for cancer patients. Methods: Machine learning models were trained usinglongitudinal retrospective EHR data from patients receiving intravenous systemic therapies for gastrointestinal cancers at the Princess Margaret Cancer Centre, aiming to predict ED visits within 30 days of each treatment. Treatments followed by ED visits within one day were excluded to ensure the system focuses on detecting early warning signs rather than imminent ED visits. The models, including tree-based methods and neural networks, were tuned with Bayesian hyperparameter optimization and calibrated using isotonic regression. A temporal split was applied to establish a held-out retrospective test cohort. The best model was silently deployed for prospective validation in patients with gastrointestinal cancer through our internally developed 'MIRA' platform that supports clinical integration. Within MIRA, the patients' EHR data with treatments scheduled the next day are extracted from the EHR system and forwarded to the model for analysis. Results: In the retrospective cohort from January 1, 2014, to December 31, 2019, 1,997 patients underwent 24,350 treatments, with 2,219 (9.11%) leading to ED visits within 30 days. The top-performing model, an extreme gradient boosting tree, achieved an area under the receiver operating characteristic curve (AUROC) of 0.68 and an area under the precision-recall curve (AUPRC) of 0.19 in the held-out test set. Although the evaluation of the system through prospective silent deployment is ongoing, here we report on patients with treatments during March 2024, with adequate 30-day follow up by April 30th, 2024. During this period, 357 patients received 676 treatments, with ED visits within 30-days following 60 (8.88%) treatments. The deployed system achieved an AUROC of 0.66 (confidence interval: 0.60-0.72) and an AUPRC of 0.22 (confidence interval: 0.15-0.32), which closely aligns with those observed during its retrospective testing. At a 10% alarm rate, model has a positive predictive value of 0.33 and sensitivity of 0.22. Conclusions: During a silent prospective deployment, our system predicted ED visits in cancer patients undergoing medical treatment. These findings indicate that the system should be integrated into the clinical workflow and combined with interventions to prevent ED visits.

Development of Machine Learning Model for Predicting Prolonged Operation Time in Lumbar Stenosis undergoing Posterior Lumbar Interbody Fusion: A Multi-center study

Background ContextLonger posterior lumbar interbody fusion (PLIF) surgeries for individuals with lumbar spinal stenosis are linked to more complications and negatively affect recovery after the operation. Therefore, there is a critical need for a method to accurately predict patients who are at risk for prolonged operation times. PurposeThis research aimed to develop a clinical model to predict prolonged operation time for patients undergoing PLIF procedures. Study Design/SettingThis study employs a machine-learning approach to analyze data retrospectively collected. Patient Sample3233 patients diagnosed with lumbar spinal stenosis (LSS) had posterior lumbar interbody fusion (PLIF) at 22 hospitals in China from January 2015 to December 2022. Outcome MeasuresThe primary outcome was operation time. Prolonged operation time defined as exceeded 75% of the overall surgical duration, which mean exceeding 240 minutes. MethodsA total of 3233 patients who underwent PLIF surgery with lumbar spinal stenosis (LSS) were divided into one training group and four test groups based on different district areas. The training group included 1569 patients, while Test1 had 541, Test2 had 403, Test3 had 351, and Test4 had 369 patients. Variables consisted of demographics, perioperative details, preoperative laboratory examinations and other Additional factors. Six algorithms were employed for variable screening, and variables identified by more than two screening methods were incorporated into the final model. In the training cohort, a 10-fold cross-validation (CV) and Bayesian hyperparameter optimization techniques were utilized to construct a model using eleven machine learning algorithms. Following this, the model was evaluated using four separate external test sets, and the mean Area Under the Curve (AUC) was computed to determine the best-performing model. Further performance metrics of the best model were evaluated, and SHapley Additive exPlanations(SHAP) were used for interpretability analysis to enhance decision-making transparency. Ultimately, an online calculator was created. ResultsAmong the various machine learning models, the Random Forest achieved the highest performance in the validation set, with AUROC scores of 0.832 in Test1, 0.834 in Test2, 0.816 inTest3, 0.822 in Test4) compared with other machine learning models. The top contributing variables were number of levels fusion, pre-APTT, weight and age. The predictive model was further refined by developing a web-based calculator for clinical application. (https://wenle.shinyapps.io/PPOT_LSS/) ConclusionsThis predictive model can facilitate identification of risk for prolonged operation time following PLIF surgery. Predictive calculators are expected to improve preoperative planning, identify patients with high risk factors, and help clinicians facilitating the improvement of treatment plans and the implementation of clinical intervention.

Deep neural network based distribution system state estimation using hyperparameter optimization

In the past decade, distribution system state estimation has become a crucial topic in power system research due to the increasing importance of distribution networks amidst the decline of centralized energy production. This paper addresses a gap in the literature regarding the application of modern hyperparameter optimization techniques in low-voltage distribution system state estimation using deep neural networks. In particular, it demonstrates the use of the Tree-structured Parzen Estimator algorithm, which is a Bayesian hyperparameter optimization method, for distribution system state estimation on real Hungarian low-voltage networks. The study uses data from four real-life low-voltage supply areas in Hungary, which were modeled to address the challenges in obtaining network information. Compared to traditional methods like the weighted least squares method, the Tree-structured Parzen Estimator algorithm significantly improves the accuracy of the voltage amplitude and angle estimations, reducing the relative error by 14–73%. Additionally, it is shown that TPE outperforms simpler methods like Random Search in hyperparameter optimization. The results also reveal connections between the distribution system size and optimal hyperparameters, such as batch size, learning rate, and hidden layer configuration. The proposed non-iterative algorithm, combined with the parallel computation capabilities of deep neural networks utilizing GPU, resulted in four orders of magnitude improvement in runtime. These advancements make the proposed approach a valuable tool for renewable energy integration planning and real-time monitoring, highlighting its potential for practical applications in the power industry.

Bayesian Optimized Machine Learning Model for Automated Eye Disease Classification from Fundus Images

Eye diseases are defined as disorders or diseases that damage the tissue and related parts of the eyes. They appear in various types and can be either minor, meaning that they do not last long, or permanent blindness. Cataracts, glaucoma, and diabetic retinopathy are all eye illnesses that can cause vision loss if not discovered and treated early on. Automated classification of these diseases from fundus images can empower quicker diagnoses and interventions. Our research aims to create a robust model, BayeSVM500, for eye disease classification to enhance medical technology and improve patient outcomes. In this study, we develop models to classify images accurately. We start by preprocessing fundus images using contrast enhancement, normalization, and resizing. We then leverage several state-of-the-art deep convolutional neural network pre-trained models, including VGG16, VGG19, ResNet50, EfficientNet, and DenseNet, to extract deep features. To reduce feature dimensionality, we employ techniques such as principal component analysis, feature agglomeration, correlation analysis, variance thresholding, and feature importance rankings. Using these refined features, we train various traditional machine learning models as well as ensemble methods. Our best model, named BayeSVM500, is a Support Vector Machine classifier trained on EfficientNet features reduced to 500 dimensions via PCA, achieving 93.65 ± 1.05% accuracy. Bayesian hyperparameter optimization further improved performance to 95.33 ± 0.60%. Through comprehensive feature engineering and model optimization, we demonstrate highly accurate eye disease classification from fundus images, comparable to or superior to previous benchmarks.

Ada-XG-CatBoost: A Combined Forecasting Model for Gross Ecosystem Product (GEP) Prediction

The degradation of the ecosystem and the loss of natural capital have seriously threatened the sustainable development of human society and economy. Currently, most research on Gross Ecosystem Product (GEP) is based on statistical modeling methods, which face challenges such as high modeling difficulty, high costs, and inaccurate quantitative methods. However, machine learning models are characterized by high efficiency, fewer parameters, and higher accuracy. Despite these advantages, their application in GEP research is not widespread, particularly in the area of combined machine learning models. This paper includes both a GEP combination model and an explanatory analysis model. This paper is the first to propose a combined GEP prediction model called Ada-XGBoost-CatBoost (Ada-XG-CatBoost), which integrates the Extreme Gradient Boosting (XGBoost), Categorical Boosting (CatBoost) algorithms, and SHapley Additive exPlanations (SHAP) model. This approach overcomes the limitations of single-model evaluations and aims to address the current issues of inaccurate and incomplete GEP assessments. It provides new guidance and methods for enhancing the value of ecosystem services and achieving regional sustainable development. Based on the actual ecological data of a national city, data preprocessing and feature correlation analysis are carried out using XGBoost and CatBoost algorithms, AdaGrad optimization algorithm, and the Bayesian hyperparameter optimization method. By selecting the 11 factors that predominantly influence GEP, training the model using these selected feature datasets, and optimizing the Bayesian parameters, the error gradient is then updated to adjust the weights, achieving a combination model that minimizes errors. This approach reduces the risk of overfitting in individual models and enhances the predictive accuracy and interpretability of the model. The results indicate that the mean squared error (MSE) of the Ada-XG-CatBoost model is reduced by 65% and 70% compared to the XGBoost and CatBoost, respectively. Additionally, the mean absolute error (MAE) is reduced by 4.1% and 42.6%, respectively. Overall, the Ada-XG-CatBoost combination model has a more accurate and stable predictive performance, providing a more accurate, efficient, and reliable reference for the sustainable development of the ecological industry.

ADNNet: Attention-based deep neural network for Air Quality Index prediction

The Air Quality Index (AQI) is a crucial indicator for assessing the degree of atmospheric pollution. Accurately forecasting AQI is notably challenging due to the unpredictable weather conditions and the intricate interactions among various pollutants. To this end, we propose a AQI prediction framework, entitled ADNNet, to develop a streamlined attention-based method for AQI prediction. Specifically, it includes a deep neural network architecture that exclusively integrates multi-layer perceptron (MLP) and attention mechanism modules. This approach involves a novel design paradigm, where data information is first fed into the MLP layer to learn global features, which are then refined by the attention module to capture local features such as time delays and periodicity. Additionally, we incorporate a Bayesian hyperparameter optimization algorithm to make the trade-off between computational time and prediction accuracy efficient. To further enhance model performance, we apply exponential smoothing (ETS) and its inverse, named inverse-ETS, boosting the model’s predictive accuracy. We leverage air pollution datasets from various cities to verify the effectiveness of the proposed ADNNet. The results demonstrate that the proposed method significantly outperforms state-of-the-art (SOTA) models like LSTM, N-BEATS, Informer, Autoformer and VMD-TCN. The compressed model with reducing the mean absolute error (MAE) by 4.17%–16.12%, decreasing the root mean squared error (RMSE) by 3.52%–12.72%, and reducing the mean absolute scaled error (MASE) by 5.26%–17.28%.The source code of our model is available at https://github.com/xiankuiwu/AQI-Attention-based-DNN.

Intelligent evaluation of interference effects between tall buildings based on wind tunnel experiments and explainable machine learning

The interference effects of taller high-rise buildings significantly impact the nearby shorter high-rise buildings in dense urban areas, potentially leading to severe wind-induced disasters. This study aims to develop a universal intelligent assessment method to predict the aerodynamic forces and wind-induced responses of buildings, considering various interfering and coupling factors. A database containing 61440 samples is established through a series of synchronous pressure measurement wind tunnel tests. The aerodynamic forces and wind-induced responses were predicted using six machine learning models under varying wind fields, interfering building locations, interfering building sizes, and dynamic characteristics of the principal building. The performance of the models was enhanced through Bayesian hyperparameter optimization and evaluated using the R2 and NRMSE. Additionally, the SHapley Additive exPlanations (SHAP) method was applied to evaluate the impact of each factor on the aerodynamic force and wind-induced response coefficients. Eight specific cases were examined using local SHAP explanations to explore the influence of seven input variables on maximum wind-induced response coefficients. Results indicated that the XGB model exhibited superior predictive performance, with R2 exceeding 0.99 in both training and testing sets. The SHAP analysis revealed significant impacts from wind direction angle, reduced wind speed, and damping ratio on wind-induced response. Consequently, irrelevant or minimally impactful input variables were removed from the XGB models. The R2 and NRMSE variation rate with reduced inputs for the optimized XGB model was less than 0.45 %. The graphical user interface (GUI) was designed can effectively guide intelligent urban planning and building design.

Neural network-based hybrid modeling approach incorporating Bayesian optimization with industrial soft sensor application

Hybrid modeling combines physical and data-driven models to improve the performance of industrial soft sensors. However, simplified physical assumptions and extensive parameter optimization increase the complexity of the model and decrease its robustness and accuracy. In this study, a novel neural network-based approach for hybrid modeling was developed, based on Runge-Kutta (RK) and Bayesian hyperparameter optimization. First, the recursive solution equation for the quality variable was derived using the master equation, which included the black-box term and RK method. Second, the numerical solution of the black-box term was inverted using the derived equation as a label for the training samples, and hyperparameters were optimized using Bayesian optimization. Finally, during the prediction phase, quality variables at subsequent times were derived by solving a recursive equation containing neural network approximations. The superiority of the proposed soft sensing method was assessed through studies on a continuous stirred tank reactor (CSTR) case and a simulated penicillin case. The experimental findings demonstrate that the proposed method is better than existing methods in terms of robustness and accuracy.

Leveraging hybrid 1D-CNN and RNN approach for classification of brain cancer gene expression

Leveraging deep learning (DL) approaches in genomics data has led to significant advances in cancer prediction. The continuous availability of gene expression datasets over the preceding years has made them one of the most accessible sources of genome-wide data, advancing cancer bioinformatics research and advanced prediction of cancer genomic data. To contribute to this topic, the proposed work is based on DL prediction in both convolutional neural network (CNN) and recurrent neural network (RNN) for five classes in brain cancer using gene expression data obtained from Curated Microarray Database (CuMiDa). This database is used for cancer classification and is publicly accessible on the official CuMiDa website. This paper implemented DL approaches using a One Dimensional-Convolutional Neural Network (1D-CNN) followed by an RNN classifier with and without Bayesian hyperparameter optimization (BO). The accuracy of this hybrid model combination of (BO + 1D-CNN + RNN) produced the highest classification accuracy of 100% instead of the 95% for the ML model in prior work and 90% for the (1D-CNN + RNN) algorithm considered in the paper. Therefore, the classification of brain cancer gene expression according to the hybrid model (BO + 1D-CNN + RNN) provides more accurate and useful assessments for patients with different types of brain cancers. Thus, gene expression data are used to create a DL classification-based- hybrid model that will hold senior promise in the treatment of brain cancer.

Detection of Bias in Machine Learning Models for Predicting Deaths Caused by COVID-19

The COVID-19 pandemic has significantly impacted global health, resulting in numerous fatalities and presenting substantial challenges to national healthcare systems due to a sharp increase in cases. Key to managing this crisis is the rapid and accurate identification of COVID-19 infections, a task that can be enhanced with Machine Learning (ML) techniques. However, ML applications can also generate biased and potentially unfair outcomes for certain demographic groups. This paper introduces a ML model designed for detecting both COVID-19 cases and biases associated with specific patient attributes. The model employs Decision Tree and XGBoost algorithms for case detection, while bias analysis is performed using the DALEX library, which focuses on protected attributes such as age, gender, race, and ethnicity. DALEX works by creating an "explainer" object that represents the model, enabling exploration of the model's functions without requiring in-depth knowledge of its workings. This approach helps pinpoint influential attributes and uncover potential biases within the model. Model performance is assessed through accuracy metrics, with the Decision Tree algorithm achieving the highest accuracy at 99% following Bayesian hyperparameter optimization. However, high accuracy does not ensure fairness, as biases related to protected attributes may still persist.

Digital-twin-driven intelligent tracking error compensation of ultra-precision machining

In ultra-precision machining (UPM), linear axis tracking affects contour accuracy and final machining quality. Traditional error modeling is complicated by the identification of numerous unknown parameters linked to nonlinear characteristics in the linear feed axes. To fill this gap, this study proposed a digital-twin-driven framework integrating the developed G-code interpreter and the deep learning model to achieve real-time tracking error compensation for UPM. To enhance the prediction accuracy of the tracking error of each axis of UPM machines, Bayesian hyperparameter optimization and feature importance analysis were conducted in the proposed TCN-BiLSTM model using high-quality training datasets from well-designed experiments. Ultimately, validation of the proposed system on a three-axis ultra-precision milling machine demonstrated its excellent performance. The experimental results showed that the optimized TCN-BiLSTM model exhibited an excellent capacity to predict the tracking error of the X-axis and Y-axis with minimal mean absolute error values of 0.000009 and 0.000023, respectively. Implementing the customized application reduced X-axis and Y-axis tracking errors by approximately 45–75% and 40–70%, respectively. This study first validates the feasibility of deep learning to improve accuracy in the UPM field, which will provide significant insight into speeding up the digitalization and intellectualization of the UPM scenario.

Bayesian optimization based extreme gradient boosting and GPR time-frequency features for the recognition of moisture damage in asphalt pavement

Moisture damage is one of the major defects in asphalt pavement, and will evolve into potholes in a short time which will affect traffic safety. Ground Penetrating Radar (GPR) is an effective non-destructive testing (NDT) method for detecting moisture damage but its data explanation replies on human experience and subjects to labor intensive. To address this issue, an automatic detection method based on extreme gradient boosting (XGBoost) combined with Bayesian hyper-parameter optimization (BHPO) was proposed to detect moisture damage area from GPR traces. High frequency GPR antenna with 2.3 GHz was used to detect the moisture damage area from simulation, laboratory and field tests, and moisture damage dataset with 7960 traces was collected. Thirty time-frequency parameters were extracted from each GPR trace, normalized to unify the three data source, and then optimized into 12 sensitive parameters by feature importance method. These 12 parameters were used to build the recognition model with XGBoost, and the model tuning parameters were optimized by BHPO. To obtain optimization model, random forest (RF) and artificial neural network (ANN) were also trained with BHPO, and compared with XGBoost model. The results indicate that performance of XGBoost model with BHPO achieves the highest accuracy and lowest time cost both in moisture damage and normal trace classification, the accuracies for moisture damage are XGBoost (96.9%) > ANN (95.6%) > RF (95.4%), respectively, and normal are XGBoost (96.5%) > RF (96.1%) and ANN (96.0%), respectively. On this basis, field tests were conducted by core samples, which verified the correct result of XGBoost model. Our method provides a swift and accurate method to locate subsurface targets directly from GPR signals.

A hybrid quantum ensemble learning model for malicious code detection

Quantum computing as a new computing model with parallel computing capability and high information carrying capacity, has attracted a lot of attention from researchers. Ensemble learning is an effective strategy often used in machine learning to improve the performance of weak classifiers. Currently, the classification performance of quantum classifiers is not satisfactory enough due to factors such as the depth of quantum circuit, quantum noise, and quantum coding method, etc. For this reason, this paper combined the ensemble learning idea and quantum classifiers to design a novel hybrid quantum machine learning model. Firstly, we run the Stacking method in classical machine learning to realize the dimensionality reduction of high-latitude data while ensuring the validity of data features. Secondly, we used the Bagging method and Bayesian hyperparameter optimization method applied to quantum support vector machine (QSVM), quantum K nearest neighbors (QKNN), variational quantum classifier (VQC). Thirdly, the voting method is used to ensemble the predict results of QSVM, QKNN, VQC as the final result. We applied the hybrid quantum ensemble machine learning model to malicious code detection. The experimental results show that the classification precision (accuracy, F1-score) of this model has been improved to 98.9% (94.5%, 94.24%). Combined with the acceleration of quantum computing and the higher precision rate, it can effectively deal with the growing trend of malicious codes, which is of great significance to cyberspace security.

Comparison of Different Machine Learning Approaches for Forecasting Stock Prices

Predicting stock prices is a crucial task that has significant implications for investment decisions, business strategies, and financial market stability. Accurate predictions can help investors make informed decisions, capture opportunities, and minimize risks. Understanding financial markets, economic data, company-specific elements, and a variety of statistical and analytical approaches are all necessary for making accurate stock price predictions. This paper employs three machine learning methods to forecast stock price data from a three-year time series dataset. Stock prediction plays a fundamental role in investment decisions, and accurate predictions are crucial. While various methods for predicting stock trends exist, different stock datasets may require different prediction models. Using three years of stock data, including gold and bitcoin, from the 2022 MCM Problem C, this study applies Support Vector Machines (SVM), Long Short-Term Memory (LSTM), and Convolutional Neural Networks (CNN) to make predictions. Bayesian hyperparameter optimization is employed, and a comparative analysis is conducted. Ultimately, the results indicate that CNN outperforms the other methods, exhibiting relatively low error metrics and a high R-squared value, with no apparent signs of overfitting.

Comparison of gridsearchcv and bayesian hyperparameter optimization in random forest algorithm for diabetes prediction

Diabetes Mellitus (DM) is a chronic disease whose complications have a significant impact on patients and the wider community. In its early stages, diabetes mellitus usually does not cause significant symptoms, but if it is detected too late and not handled properly, it can cause serious health problems. To overcome these problems, diabetes detection is one of the solutions used. In this research, diabetes detection was carried out using Random Forest with gridsearchcv and bayesian hyperparameter optimization. The research was carried out through the stages of study literature, model development using Kaggle Notebook, model testing, and results analysis. This study aims to compare GridSearchCV and Bayesian hyperparameter optimizations, then analyze the advantages and disadvantages of each optimization when applied to diabetes prediction using the Random Forest algorithm. From the research conducted, it was found that GridSearchCV and Bayesian hyperparameter optimization have their own advantages and disadvantages. The GridSearchCV hyperparameter excels in terms of accuracy of 0.74, although it takes longer for 338,416 seconds. On the other hand, Bayesian hyperparameter optimization has a lower accuracy rate than GridSearchCV optimization with a difference of 0.01, which is 0.73 and takes less time than GridSearchCV for 177,085 seconds.

Application of Ensemble Learning and Feature Selection Models in Energy Expenditure Estimation of Fitness Tracking Devices

The use of fitness tracking devices to monitor energy expenditure (EE) is becoming increasingly popular among consumers. However, the accuracy of EE estimation still needs to be improved due to the lack of accurate data filtering and more efficient data analysis models. To address this issue, we designed an energy expenditure estimation (EEE) model based on smart tracking devices. This paper proposes an algorithm for data reconstruction called PC-DF-RFECV: firstly, Pearson correlation analysis (PC) is used for initial data screening, then Feature Derivation (DF) is applied to reconstruct new features, and finally, the Recursive Feature Elimination algorithm based on cross-validation (RFECV) is employed to select the new features. Furthermore, the EEE model adopts a weighted average (Wei-Voting) strategy to enhance the robustness of the model by integrating the predictions of multiple base learners (XGBoost, LightGBM, Random Forest). During algorithm optimization, a Bayesian hyperparameter optimization technique is employed to fine-tune the model's hyperparameters. Empirical findings indicate that the EEE model, when applied to the dataset collected from motion tracking devices, attains an score of 0.9985, of 1.033, and of 0.865. These results demonstrate superior performance compared to current state-of-the-art research and conventional algorithms including random forests, gradient boosting, and bagging.

Enhanced Two-Stream Bayesian Hyper Parameter Optimized 3D-CNN Inception-v3 Based Drop-ConvLSTM2D Deep Learning Model for Human Action Recognition

Human Action Recognition (HAR) has grown to be the toughest and most attractive concern in the domains of computer vision, communication between a person and the surroundings, and video surveillance. In variation to the conventional methods that usually make use of the Long Short Term Memory model (LSTM) for training, this work designed dropout variant Drop-ConvLSTM2D, to provide more effectiveness in regularization for deep Convolution Neural Networks (CNNs).In addition, to speed up the runtime performance of the Deep Learning model, Bayesian Hyper Parameter Optimization (BHPO) is also introduced to autonomously optimize, the hyperparameters of the trained architecture. In this study, a two-stream Bayesian Hyper Parameter optimized Drop-ConvLSTM2D model is designed for HAR to overcome the current research deficiencies. In one stream, an Inception-v3 model extracts the temporal characteristics from the optical frames which are generated through the dense flow process. In another stream, a 3D-CNN involves the mining of the spatial-temporal characteristics from the RGB frames. Finally, the features of Inception-v3 and 3D-CNN are fused using which the Drop-ConvLSTM2D model is trained to recognize human behavior. On perceptive public video datasets UCF-101, and HMDB51, the quantitative assessments are conducted on the Drop-ConvLSTM2D BHPO model. For all hyperparameters, the built model explicitly obtains optimized values in this process, which can save time and improve performance. The experimental outcome shows that with a precision of at least 3%, the designed model beats the traditional two-stream model.

Multi-source driven estimation of earthquake economic losses: A comprehensive and interpretable ensemble machine learning model

Rapid and precise quantification of economic losses post-earthquake is critical for crafting informed disaster management strategies by governmental and insurance entities. This study introduces an ensemble model, constituted by Support Vector Machines (SVM) and Extreme Gradient Boosting (XGBoost), tailored for quick and interpretative prediction of GDP-related seismic loss assessments, with Sichuan Province serving as the empirical backdrop. The ensemble approach, pre-trained for expediency, draws on a knowledge-driven selection of 30 features that span seismic risks, socio-economic variables, and exposure factors. Enhanced by feature scaling and Bayesian hyperparameter optimization, the model's efficacy is validated through 6 metrics such as R-Squared, MSE, and MAE etc. The incorporation of SHAP values further unravels the model's decision-making, providing transparency to the often opaque computations of machine learning. This methodology offers a scalable, interpretable framework that equips stakeholders with timely and accurate insights for risk mitigation, ultimately strengthening earthquake resilience.

Hybrid Neural Networks for Enhanced Predictions of Remaining Useful Life in Lithium-Ion Batteries

With the proliferation of electric vehicles (EVs) and the consequential increase in EV battery circulation, the need for accurate assessments of battery health and remaining useful life (RUL) is paramount, driven by environmentally friendly and sustainable goals. This study addresses this pressing concern by employing data-driven methods, specifically harnessing deep learning techniques to enhance RUL estimation for lithium-ion batteries (LIB). Leveraging the Toyota Research Institute Dataset, consisting of 124 lithium-ion batteries cycled to failure and encompassing key metrics such as capacity, temperature, resistance, and discharge time, our analysis substantially improves RUL prediction accuracy. Notably, the convolutional long short-term memory deep neural network (CLDNN) model and the transformer LSTM (temporal transformer) model have emerged as standout remaining useful life (RUL) predictors. The CLDNN model, in particular, achieved a remarkable mean absolute error (MAE) of 84.012 and a mean absolute percentage error (MAPE) of 25.676. Similarly, the temporal transformer model exhibited a notable performance, with an MAE of 85.134 and a MAPE of 28.7932. These impressive results were achieved by applying Bayesian hyperparameter optimization, further enhancing the accuracy of predictive methods. These models were bench-marked against existing approaches, demonstrating superior results with an improvement in MAPE ranging from 4.01% to 7.12%.