MSE Values Research Articles

Interpretable machine learning-based analysis of coal fly ash flotation by conditioning process intensification

ABSTRACT Conditioning process intensification is necessary for improving the fly ash flotation effect. Understanding the relationship between conditioning parameters and flotation results is crucial for optimizing the conditioning-flotation process. This study developed a machine learning framework for predicting the yield, loss on ignition, and removal rate of unburned carbon using data from conditioning-flotation experiments. Feature variables of conditioning speed, impeller type, stirred tank diameter, collector dosage, and frother dosage were employed to develop five machine learning models. After tuning the hyper-parameters with GridSearchCV technique, it was found that Random Forest Regression model, eXtreme Gradient Boosting Regression model, and Gradient Boosting Regression model demonstrated good agreements between the predicted data and test data for the yield, loss on ignition, and removal rate of unburned carbon. The maximum R2 values were 0.9184, 0.8339, and 0.9434, respectively, while the minimum MSE values were 3.30, 5.82, and 20.59, respectively. The interpretable analysis using SHapley Additive exPlanations (SHAP) method indicated that conditioning speed was the most significant feature variable affecting the yield, loss on ignition, and removal rate of unburned carbon, while impeller type had a secondary impact. This investigation is expected to reveal the important role of conditioning process intensification in optimizing fly ash flotation.

MFR-Net: A multi-feature fusion phase unwrapping method for different speckle noises

Phase unwrapping is a crucial step in laser interferometry for obtaining accurate physical measurement of object. To reduce the impact of speckle noise on wrapped phase during actual measurement and improve the subsequent measurement accuracy, a multi-feature fusion phase unwrapping method for different speckle noises named MFR-Net is proposed in this paper. The network is composed of a front-end multi-module filter processing layer and a back-end network with dilated convolution and coordinate attention mechanism. By reducing random phase differences introduced by different levels of noise, the network enhances its capability to extract spatial features such as gradient information between pixels under speckle noise, so that it successfully unwraps the wrapped phase with different speckle noises and accurately recovers the real phase information. Taking the wrapped phases with multiplicative speckle noise and additive random noise as dataset, the results of ablation and comparison experiments show that the MFR-Net has superior unwrapped results. Under three different levels of speckle noise, the average values of MSE, SSIM, PSNR and AU for MFR-Net are at least improved by 84.80 %, 10.99 %, 29.00 % and 7.72 %, respectively, compared to PDVQG, TIE, DLPU and VURNet algorithms. When the standard deviation of speckle noise varies continuously in the range [1.0, 2.0], the average values of four indexes reaches 0.12 rad, 0.91, 31.80 dB and 99.96 %, respectively, indicating the stronger robustness of MFR-Net. In addition, the phase step unwrapping is performed by MFR-Net. Compared to DLPU and VURNet, MFR-Net method reduced MSE by 80 % and 87.35 %, respectively, demonstrating the outstanding generalization capability. The proposed MFR-Net can realize the correct phase unwrapping under different speckle noises. It may be applied in laser interferometry applications such as digital holography and interferometric synthetic aperture radar.

Forecasting of meteorological drought using ensemble and machine learning models

This study highlights drought forecasting for understanding the semi-arid area in India, where drought phenomena play vital role in the irrigation, drinking water supplies, and sustaining the ecological with economic balance for every nation. Therefore, drought forecasting is important for the future drought planning based on the machine learning (ML) models. Hence, The Standardized Precipitation Index (SPI) at 3- and 6-month periods have been selected and used for future drought forecasting scenarios in area. The combinations of ten inputs SPI-1- and SPI-10 were used for predicting modeling for SPI-3 and SPI-6 timescales, that modeling developed based on the historical SPI datasets from 1989 to 2019 years. The SPI-3 and SPI-6 maximum and minimum values are shown SPI-3 (2.03 and -5.522) and SPI-6 (1.94 and -6.93). The SPI is a popular method for estimating the drought analysis and has been used everywhere at global level. The developed models have been compared with each other, with the best combination of input variables selected using subset regression models and sensitivity studies. After that, the active input parameters were used for forecasting of SPI-3 and SPI-6 values to understanding of drought in semi-arid area. The finest input variables combination have been used in the Ml models and established the novel five models such as robust linear regression, bagged trees, boosted trees, support vector regression (SVM-Linear), and Matern Gaussian Process Regression (Matern GPR) models. Such kind of models first time has been applied for the forecasting of future drought conditions. Whole models were fine and improved modeling by using hyperparameters tuning, bagging, and boosting models. Entire ML models’ accuracy was compared using different statistical metrics. Compared with five ML models accuracy, we have found that the Matern GPR model better accuracy than other ML models. The best model accuracy is R2 = 0.95 and 0.93, RMSE, MSE, MAE, MARE, and NSE values, respectively, for predicting SPI-3 and SPI-6 values in the area. Therefore, the Matern GPR model was identified as the finest ML algorithm for predicting SPI-3 and SPI-6 associated with other algorithms. This research demonstrates the Matern GPR model's efficacy in predicting multiscale SPI-3 and SPI-6 under climate variations. It can be helpful in soil and water resource conservation planning and management and understanding droughts in the entire basin areas of the country India.

Computational fluid dynamics (CFD)- deep neural network (DNN) model to predict hydrodynamic parameters in rectangular and cylindrical bubble columns

Bubble columns are omnipresent in the chemical, bio-chemical, petrochemicals, petroleum industries, but their design and scale-up is complex owing to its complex hydrodynamics. Liquid velocity and gas holdup is one of the critical hydrodynamic parameters which effects the mixing, heat and mass transfer in bubble columns. CFD is widely recognized as a powerful tool for estimating critical hydrodynamic parameters but requires significant computational resources, time and expertise. These limitations restrict its practical use in hydrodynamic simulations that need real-time processing involving large-scale simulations of bubble columns. To overcome these limitations, CFD-DNN model is developed to predict the time averaged gas holdup and axial liquid velocity at various operating conditions. The DNN model was trained using CFD data that was produced for rectangular (with dimensions L=0.2 m, W=0.05 m, H=1.2 m) and cylindrical (with a diameter of 0.19 m) bubble columns. The data covers a range of operating conditions and various flow regimes. The superficial gas velocity for the rectangle column was selected at 1.33 and 7.3 mm/s, whereas for the cylindrical bubble column, it was fixed at 0.02 and 0.12 m/s. The CFD-DNN model was validated against the experimental and the CFD data from the literature. Further, the model was tested for new data that the CFD-DNN model has not seen with existing literature and showed good agreement with their data and it reflects the excellent generalization ability of the model. The proposed CFD-DNN approach improves current CFD models by providing shorter computing time, decreasing computational expenses, and reducing the expertise in CFD simulations. The accuracy of the developed CFD-DNN model was evaluated using different metrics for gas holdup and axial liquid velocity. For rectangular bubble columns, the model achieved MSE of 0.0001 for gas holdup and 0.0007 for axial liquid velocity. Similarly, for cylindrical bubble columns, the MSE values were 0.0009 for gas holdup and 0.0006 for axial liquid velocity.

Different estimation techniques and data analysis for constant-partially accelerated life tests for power half-logistic distribution

Partial accelerated life tests (PALTs) are employed when the results of accelerated life testing cannot be extended to usage circumstances. This work discusses the challenge of different estimating strategies in constant PALT with complete data. The lifetime distribution of the test item is assumed to follow the power half-logistic distribution. Several classical and Bayesian estimation techniques are presented to estimate the distribution parameters and the acceleration factor of the power half-logistic distribution. These techniques include Anderson–Darling, maximum likelihood, Cramér von-Mises, ordinary least squares, weighted least squares, maximum product of spacing and Bayesian. Additionally, the Bayesian credible intervals and approximate confidence intervals are constructed. A simulation study is provided to compare the outcomes of various estimation methods that have been provided based on mean squared error, absolute average bias, length of intervals, and coverage probabilities. This study shows that the maximum product of spacing estimation is the most effective strategy among the options in most circumstances when adopting the minimum values for MSE and average bias. In the majority of situations, Bayesian method outperforms other methods when taking into account both MSE and average bias values. When comparing approximation confidence intervals to Bayesian credible intervals, the latter have a higher coverage probability and smaller average length. Two authentic data sets are examined for illustrative purposes. Examining the two real data sets shows that the value methods are workable and applicable to certain engineering-related problems.

Peramalan Cadangan Devisa Menggunakan Metode Double Smoothing Exponential dan Metode Fuzzy Time Series

This study aims to forecast the amount of foreign exchange reserves using two methods, namely Double Smoothing Exponential and Fuzzy Time Series. This foreign exchange reserve data is time series data, namely monthly data starting from January 2017 to February 2024 with a total of 86 monthly data. Furthermore, to see the level of accuracy of the forecast results, the Mean Absolute Percentage Error (MAPE) and MSE values ​​are used where the smaller the error value produced, the more accurate the forecast results from the Double Smoothing Exponential Holt and Fuzzy Time Series Cheng methods. The results of forecasting the amount of foreign exchange reserves in March 2024 using the Holt method are 144434.19 and for the Cheng method are 140008.032. The double exponential smoothing method has an MSE value of 8625715.609 and a MAPE value of 1.69%. Meanwhile, with the fuzzy time series method, the MSE value is 11651319.400 and the MAPE value is 1.96%. Forecasting using these two methods has a very good level of accuracy, because it has a MAPE value below 10%. Based on the results of this analysis, it can be seen that the double exponential smoothing method from Holt has a smaller forecast error size than the Cheng fuzzy time series method. Thus, it can be concluded that the Double Exponential Smoothing method is a better method for predicting the amount of foreign exchange reserves.

Data-driven performance analysis of an active chilled beam air conditioning system: A machine learning approach for energy efficiency and predictive maintenance

This study aims to enhance energy performance estimation and maintenance planning for office buildings in Tehran. Machine learning and data mining techniques were employed, utilizing Support Vector Machines (SVMs), Artificial Neural Networks (ANNs), Narrow Neural Networks, and optimizable SVMs. A comprehensive dataset of 2500 experimental samples was utilized with five input parameters (building size, occupancy patterns, environmental conditions, and ACB system parameters) and one output parameter (electricity consumption). GPR models performed better in estimating electricity consumption, achieving a Root Mean Square Error (RMSE) of 297.01 and an impressive R-squared value of 0.99. To optimize energy performance and design more energy-efficient buildings, this model accurately predicts energy consumption. Additionally, this study addressed maintenance planning for ACB systems, simulating maintenance issues with 2500 synthetic samples. The dataset included building age, size, number of floors, usage intensity, previous maintenance history, location, and environmental factors. Maintenance time was estimated using these features, and XGBoost was employed to predict ACB system maintenance time. Low MSE, RMSE, and R-squared values in the XGBoost model allow for accurate predictions. Using this model, facility managers can proactively plan maintenance activities, allocate resources, and minimize downtime. Through proactive measures, ACB performance and maintenance costs are optimized. The study accurately predicts ACB system energy consumption and maintenance time using the rational quadratic GPR model, applicable in Tehran and beyond. Additionally, MATLAB version R2022b was used for prediction, while DesignBuilder and EnergyPlus were utilized for technical analysis.

GREEN BOND INDEX PRICE FORECASTING: COMPARATIVE ANALYSIS OF MACHINE LEARNING MODELS

Today, green financial products have garnered recognition and are consequently regarded as alternative assets. Green bonds exclusively allocate their funds to environmentally sustainable initiatives. Green bonds facilitate companies in enhancing both their financial and environmental performance by promoting innovations stemming from green initiatives and long-term green investments. This study focused on predicting the price of the green bond index in Japan. The input factors for price prediction in Japan include Nikkei225, USD/JPY, and crude oil price assets, which are seen as alternative investment options for Japanese investors. The study utilized a dataset spanning 693 days, from 06.05.2021 to 02.05.2024. The acquired data is partitioned into two distinct sets: one for training and one for testing. 80% of the data was allocated for training purposes, while the remaining 20% was designated for testing. The study utilized various prediction approaches including RF, MLP, GBR, XGBoost, LSTM, SVR, Catboost, and Linear Regression. The performance of these models was compared using evaluation metrics such as MSE, RMSE, MAE, MAPE, and R2 values. The research revealed that the GBR model exhibited the highest performance on the training data set, whereas the XGBoost and RF models yielded superior prediction results on the test data set.

Uncertainty analysis of photovoltaic power generation system and intelligent coupling prediction

Accurate prediction of photovoltaic power generation is essential to promoting the active consumption and low-carbon protection. The complex uncertainty of the photovoltaic system itself leads to the deviation in the photovoltaic power prediction. Therefore, we propose a new prediction model for coupled intelligence optimization. First, the photovoltaic power is decomposed into effective mode components using VMD optimized by GWO. Statistical techniques were used to analyze multidimensional uncertainty and extract features, then, optimize the performance of the coupled model. Second, the Zebra optimization (ZOA) establishes an appropriate balance between exploration and utilization to achieve the optimization of the model parameters. In addition, the CNN is used to extract complex features and enhance the correlation between input values and output values. Finally, the power was predicted using the BiLSTM. The results show that applying the statistical technique to the coupled prediction model not only reveals the uncertainty of photovoltaic systems but reduces the prediction error. Among them, the R2 increased by 0.42 %, the values of MAPE, MSE, RMSE, and MAE were reduced to different degrees. It can better optimize the allocation and reasonable consumption of renewable energy, which provides the decision basis for the adjustment of renewable energy structure.

Comparative Study in Controlling Outliers and Multicollinearity Using Robust Performance Jackknife Ridge Regression Estimator Based on Generalized-M and Least Trimmed Square Estimator

Regression analysis is one of the statistical methods used to determine the causal relationship between one or more explanatory variables to the affected variable. The problem that often occurs in regression analysis is that there are multicollonity and outliers. To deal with such problems can be solved using ridge regression analysis and robust regression. Ridge regression can solve the problem of multicollinearas by assigning a constant k to the matrix Z′Z. But in this method the resulting bias value is still high, so to overcome this problem, the jackknife ridge regression method is used. Meanwhile, to overcome outliers in the data using robust regression methods which have several estimation methods, two of which are the Generalized-M (GM) estimator and the Least Trimmed Square (LTS) estimator. The aim of the study is to solve the problem of multicollinearity and outliers simultaneously using robust jackknife ridge regression method with GM estimators and LTS estimators. The results showed that the robust ridge jackknife regression method with LTS estimator can control multicollinearity and outliers simultaneously better based on MSE, AIC and BIC values compared to the robust ridge jackknife regression method with GM estimators. This is indicated by the value MSE = -6.60371, AIC = 75.823 and BIC = 81.642 on LTS estimators that are of lower value than GM estimators.

Predicting PM2.5 concentrations in Ba Ria - Vung Tau province using an artificial neural network model

Abstract Air pollution has increased rapidly and significantly recently, particularly in big cities. Various methods for predicting air pollution are available, including traditional air quality models, statistical techniques, and artificial intelligence. In this study, the author developed a model using a Feed Forward Neural Network with multivariate statistical methods to predict air pollution. Data from three automated air monitoring locations in Ba Ria-Vung Tau province were gathered between 2020 and 2022 to forecast the concentration of PM2.5. The results demonstrated that the FFNN model with an I(6)-HL(5)-O(1) structure outperformed other models in predicting PM2.5 concentration. The training, validation, and testing phases yielded mean squared error values of 9.2×10−6, 8.2×10−6, and 8.6×10−6, respectively. The regression coefficient obtained consistently high values across a range of experiments (above 0.99). The MSE value of the FFNN model of the prediction set was lower than that of the NSE value, which was higher than those obtained from the multiple linear regression.

Application of improved harmonic Poisson segmented regression model in evaluating the effectiveness of Kala-Azar intervention in Yangquan City, China.

The Centre for Disease Control and Prevention in Yangquan, China, has taken a series of preventive and control measures in response to the increasing trend of Kala-Azar. In response, we propose a new model to more scientifically evaluate the effectiveness of these interventions. We obtained the incidence data of Kala-Azar from 2017 to 2021 from the Centre for Disease Control and Prevention (CDC) in Yangquan. We constructed Poisson segmented regression model, harmonic Poisson segmental regression model, and improved harmonic Poisson segmented regression model, and used the three models to explain the intervention effect, respectively. Finally, we selected the optimal model by comparing the fitting effects of the three models. The primary analysis showed an underlying upward trend of Kala-Azar before intervention [incidence rate ratio (IRR): 1.045, 95% confidence interval (CI): 1.027-1.063, p < 0.001]. In terms of long-term effects, the rise of Kala-Azar slowed down significantly after the intervention (IRR:0.960, 95%CI:0.927-0.995, p = 0.026), and the risk of Kala-Azar increased by 0.3% for each additional month after intervention (β1 + β3 = 0.003, IRR = 1.003). The results of the model fitting effect showed that the improved harmonic Poisson segmental regression model had the best fitting effect, and the values of MSE, MAE, and RMSE were the lowest, which were 0.017, 0.101, and 0.130, respectively. In the long term, the intervention measures taken by the Yangquan CDC can well curb the upward trend of Kala-Azar. The improved harmonic Poisson segmented regression model has higher fitting performance, which can provide a certain scientific reference for the evaluation of the intervention effect of seasonal infectious diseases.

Bi-Directional Prediction Model for Hot Pressing Production Parameters and Quality of High-Performance Bamboo-Based Fiber Composites Based on cHGWOSCA-SVR

In the hot press process of high-performance bamboo-based fiber composites, there is a highly nonlinear relationship between the production parameters of hot press and the quality parameters of the finished boards. Consequently, it is challenging to accurately predict the quality of the boards based on the given production parameters, and it is equally difficult to preset the production parameters to achieve the desired board quality. The current approach relies on manual experience, which may result in subpar board quality and material waste. To address these issues, this paper proposes a bi-directional prediction model based on cHGWO-SCA-SVR, using the collaboration-based hybrid GWO-SCA optimizer to optimize the relevant parameters of the SVR, and then accurately predicting the production parameters and the quality of the finished boards in both directions. Finally the cHGWO-SCA-SVR prediction model achieves an average of 0.9591 for the forward prediction model and lower MAE and MSE values compared to other models; for the reverse prediction model, it attains an average of 0.9553 and lower MAE and MSE values compared to other models. The results demonstrate the superiority of the cHGWO-SCA-SVR prediction model in comparison with other existing models, proving its significance in guiding the production of high-performance bamboo-based fiber composites by hot compression.

Time series forecasting of weight for diuretic dose adjustment using bidirectional long short-term memory

Loop diuretics are prevailing drugs to manage fluid overload in heart failure. However, adjusting to loop diuretic doses is strenuous due to the lack of a diuretic guideline. Accordingly, we developed a novel clinician decision support system for adjusting loop diuretics dosage with a Long Short-Term Memory (LSTM) algorithm using time-series EMRs. Weight measurements were used as the target to estimate fluid loss during diuretic therapy. We designed the TSFD-LSTM, a bi-directional LSTM model with an attention mechanism, to forecast weight change 48 h after heart failure patients were injected with loop diuretics. The model utilized 65 variables, including disease conditions, concurrent medications, laboratory results, vital signs, and physical measurements from EMRs. The framework processed four sequences simultaneously as inputs. An ablation study on attention mechanisms and a comparison with the transformer model as a baseline were conducted. The TSFD-LSTM outperformed the other models, achieving 85% predictive accuracy with MAE and MSE values of 0.56 and 1.45, respectively. Thus, the TSFD-LSTM model can aid in personalized loop diuretic treatment and prevent adverse drug events, contributing to improved healthcare efficacy for heart failure patients.

Perbandingan Analisis Komponen Utama Robust Minimum Covarian Determinant dengan Least Trimmed Square pada Data Produk Domestik Regional Bruto

Regression analysis is a method to examine the relationship between variables and determine their influence. However, the problem of multicollinearity often arises in linear regression analysis and can cause interpretation problems. To handle multicollinearity, Principal Component Analysis (PCA) is used. However, this method has a weakness when the data contains outliers. Therefore, it was developed into robust PCA using the Minimum Covariance Determinant (MCD) method and the Least Trimmed Square (LTS) estimation method. This study uses Gross Regional Domestic Product data in Indonesia in 2020, which has problems with multicollinearity and outliers. This data is modeled using two robust PCA methods, namely MCD and LTS. The robust PCA model with MCD has an adjusted value of 87.87% and an MSE value of 0.0700. However, in the robust PCA regression model with LTS, the adjusted value is 98.93% and the MSE value is 0.0325. Thus, the effective method in handling multicollinearity and outliers is the LTS method because it shows better results.

Estimasi Parameter Regresi Ridge Robust pada Data Profil Kesehatan Sulawesi Selatan

ABSTRACT Multicollinearity is one of the assumption violations in regression analysis. The existence of multicollinearity causes the standard error to increase. Ridge regression is one of the regression analysis approaches that can overcome this problem. Besides multicollinearity, another problem that often occurs is outliers. The existence of outliers causes the data not to be normally distributed. Ridge Robust Least Trimmed Square Regression is a method that can be used to overcome multicollinearity and outlier problems in the data simultaneously in the regression analysis model. The purpose of this study was to obtain the estimation results of the least trimmed square ridge robust regression model on the Health Profile data of South Sulawesi in 2017. From the results and discussion it was found that the least trimmed square ridge robust regression method has an Rsquare value or ?2 which is 88% and an MSE value 1.96, thus indicating that the ridge robust least trimmed square model fits better in dealing with data containing multicollinearity and outliers. Keywords: Robust Ridge Regression, Least Trimmed Square, Multicollinearity, Outlier, Infant Mortality Rate.

Design of Immersive VR Tourism Analysis Model Based on Fuzzy Logic Algorithm

Immersive virtual reality (VR) is a technology that transports users into fully immersive digital environments, often through the use of specialized headsets and sensory equipment. A three-dimensional environment, immersive VR offers users an unparalleled sense of presence and interaction, enabling them to explore and interact with virtual worlds as if they were physically present. This paper develops an effective immersive Virtual Reality (VR) model for tourism. The proposed model uses the fuzzy-based logic rules for tourism in VR for the classification with the Backpropagation Feedforward Neural Network (BFNN). Through the developed model efficacy of BFNN-based algorithms in accurately classifying diverse virtual environments and detecting edges with precision. The analysis of the results stated that the through BFNN model MSE and PSNR value is achieved with the value of 0.007 and 28.9 respectively. With the developed model significant classification is achieved with the BFNN model value for the exploration, cultural heritage, adventure, Urban exploration, and Relaxation. Additionally, comparative analyses demonstrate the superiority of BFNNs over alternative classification models, underscoring their effectiveness in accurately categorizing immersive tourism experiences. These findings stated that the advancement of immersive VR technology also offers practical insights for optimizing computational algorithms in immersive tourism applications. The potential of BFNNs redefine the landscape of immersive VR tourism, delivering captivating and personalized virtual experiences that elevate user engagement and satisfaction to unprecedented levels.

Predicting Weld Quality in Duplex Stainless Steel Butt Joints During Laser Beam Welding: A Hybrid DNN-HEVA Approach

Duplex stainless steel (DSS) welding is critical for producing structures and components in a variety of industries. Because traditional optimization approaches cannot manage the complexity of welding, additional solutions must be investigated. This study addresses the need for a systematic exploration of the welding parameters affecting the quality and performance of DSS butt joints, specifically employing fiber laser welding. This study focuses on butt joints in UNS S32304 and 304L using a fiber laser. Welding process parameters were systematically varied, including laser power, scanning speed, beam diameter, and focal position. To efficiently examine the parameters in the design space, the response surface methodology (RSM) with Box–Behnken design (BBD) was used. The combination of the specific parameter values used in the 7th run (Laser Power: 1600 W, Scanning Speed: 2000 mm/min, Beam Diameter: 1.5 [Formula: see text]m, Focal Position: 30 mm) led to better yield point stress (YPS), maximum tensile stress (MTS), break stress (BS) and reduction in area (RA). ANOVA and lack of fit analysis confirmed the significance of the experiment. In particular, the selected welding parameters significantly influenced all responses. A hybrid machine learning method of deep neural network (DNN) and human eye vision algorithm (HEVA) was used for predicting weld quality during laser beam welding (LBW). The hybrid DNN-HEVA consistently outperformed other optimization methods (RSM-BBD, FNN, CNN, and RNN) across key welding quality parameters. High [Formula: see text]2 values (0.9922, 0.9962, 0.9983, and 0.9907) indicated a strong correlation between predicted and actual values for all calculated responses. Lower RMSE, MSE, and MAE values were also highlighted in the precision of predicted values to the experimental data.

Cryptocurrencies Price Estimation Using Deep Learning Hybride Model of LSTM-GRU

One of the financial assets in currency exchange is now cryptocurrency. The public is drawn to cryptocurrency trading because it is considered a lucrative form of investing. For cryptocurrency investors to maximize their earnings, accurate price forecasting is crucial. As price forecasting involves time series analysis, a hybrid deep learning model is suggested to project cryptocurrency prices in the future. Long Short-Term Memory and Gated Recurrent Unit (LSTM-GRU) networks are integrated into the hybrid model. Three cryptocurrency datasets are evaluated using the suggested hybrid model: Ethereum, Ripple, and Bitcoin. According to experimental results, the suggested LSTM-GRU model may provide the lowest MSE and RMSE values on the Bitcoin dataset (0.0611 and 0.2472), the Ethereum dataset (0.0369 and 0.19222), and the Ripple dataset (0.0006 and 0.0247).

Enhancing Energy Consumption Prediction in Smart Homes: A Convergence-Aware Federated Transfer Learning Approach

Achieving accurate energy consumption prediction can be challenging, particularly in residential buildings, which experience highly variable consumption behavior due to changes in occupants and the construction of new buildings. This variability, combined with the potential for privacy breaches through conventional data collection methods, underscores the need for novel approaches to energy consumption forecasting. The proposed study suggests a new approach to predict energy consumption, utilizing Federated Learning (FL) to train a global model while ensuring local data privacy and transferring knowledge from information-rich to information-poor buildings. The proposed method learns the transferable knowledge from the source building without any privacy leakage and utilizes it for target buildings. Since the performance of the global model could be negatively affected by some participating nodes with poor performance due to noisy or limited data, we propose a client selection strategy on the server based on the normal distribution for choosing the best possible participants for the global model. Our method enables clients to participate selectively in the aggregation procedure to avoid model divergence due to poor performance. The proposed model is evaluated and conducts in-depth analyses of energy consumption patterns. We validate the performance by comparing its MAE, MSE, and R2 values to those of existing traditional and ensemble models. Our findings indicate that the proposed FL-based model with selective client participation outperforms its counterpart methods regarding predictive accuracy and robustness. The source code is available on Github