Search

Quality testing and monitoring advancements have allowed modern production processes to achieve extremely low failure rates, especially in the era of Industry 4.0. Such processes are known as high‐yield processes, and their data set consists of an excess number of zeros. Count models such as Poisson, Negative Binomial (NB), and Conway‐Maxwell‐Poisson (COM‐Poisson) are usually considered good candidates to model such data, but the excess zeros are larger than the number of zeros, which these models fit inherently. Hence, the zero‐inflated version of these count models provides better fitness of high‐quality data. Usually, linearly/non‐linearly related variables are also associated with failure rate data; hence, regression models based on zero‐inflated count models are used for model fitting. This study is designed to propose deep learning (DL) based control charts when the failure rate variables follow the zero‐inflated COM‐Poisson (ZICOM‐Poisson) distribution because DL models can detect complicated non‐linear patterns and relationships in data. Further, the proposed methods are compared with existing control charts based on neural networks, principal component analysis designed based on Poisson, NB, and zero‐inflated Poisson (ZIP) and non‐linear principal component analysis designed based on Poisson, NB, and ZIP. Using run length properties, the simulation study evaluates monitoring approaches, and a flight delay application illustrates the implementation of the research. The findings revealed that the proposed methods have outperformed all existing control charts. [ABSTRACT FROM AUTHOR]

This research investigates machine learning models for predicting mental health consequences using survey data. The study employs a two-phase approach first, it utilizes TensorFlow for initial Deep Neural Network (DNN) model building, and then it applies Random Forest (RF), Naive Bayes classifier, and decision tree methods for comparative analysis. The DNN model demonstrates strong performance, achieving high accuracy in mental health prediction. Metrics such as testing time, precision, mean absolute error, and accuracy are compared to provide insight into the advantages and disadvantages of each model. While the DNN model excels in accuracy and precision, other models offer trade-offs in computational efficiency. The results clarify the role of machine learning in mental wellness evaluation and intervention, providing guidance for further research and real-world applications. This research enhances the discourse on predictive modeling for mental health outcomes, facilitating advancements in leveraging machine learning to improve mental health assessment and intervention strategies. [ABSTRACT FROM AUTHOR]

Developed countries have substantial investments in renewable energy currently, particularly Photovoltaics (PV), for achieving net-zero emissions. But PV generation is highly volatile and hence achieving supply-demand balance is challenging. Robust forecasting models will help PV integration and penetration into the grid, making sure that there is an adequate supply to match the demand, ensuring reliability and stability of power systems. In this paper, a deep learning model is developed for PV generation multistep forecasting using a small subset of weather variables with a 15-minute resolution, with very low computation time. The forecasts very closely align with the actual generation, with a Normalized Mean Absolute Error (nMAE) of 0.04, much less than 1 kWh in terms of error in forecast generation. Direct and multioutput forecasting are combined here. Comparisons with Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU) show performance improvement, by ∼15% compared to LSTM and ∼17% compared to GRU in terms of average nMAE. The model can be used in urban environments for short term forecasting. Also, if an accurate forecast is available, PV asset owners can plan their generation better when they export power back into the grid, make better bids in the energy markets, increase their revenues and eventually increase the share of renewables in the energy market.

Principal component analysis (PCA) is essential for diminishing the number of dimensions across various fields, preserving data integrity while simplifying complexity. Eigenfaces, a notable application of PCA, illustrates the method's effectiveness in facial recognition. This paper introduces a novel PCA approximation technique based on maximizing distance and compares it with the traditional eigenfaces approach. We employ several image quality metrics including Euclidean distance, mean absolute error (MAE), peak signal-to-noise ratio (PSNR), signal-to-noise ratio (SNR), and structural similarity index measure (SSIM) for a quantitative assessment. Experiments conducted on the Brazilian FEI database reveal significant differences between the approximated and classical eigenfaces. Despite these differences, our approximation method demonstrates superior performance in retrieval and search tasks, offering faster and parallelizable implementation. The results underscore the practical advantages of our approach, particularly in scenarios requiring rapid processing and expansion capabilities. [ABSTRACT FROM AUTHOR]

Speech audiometry is a vital tool in assessing an individual's ability to perceive and comprehend speech, traditionally requiring specialized testing that can be time-consuming and resource -intensive. This paper approaches a novel use of deep learning to predict speech audiometry using pure-tone audiometry (PTA) data. By utilizing PTA data, which measure hearing sensitivity at specific frequencies, we aim to develop a model that can bypass the need for direct speech testing. This study investigates two neural network architectures: a multi-layer perceptron (MLP) and a one-dimensional convolutional neural network (1D-CNN). These models are trained to predict key speech audiometry outcomes, including speech recognition thresholds and speech discrimination scores. To evaluate the effectiveness of these models, we employed two key performance metrics: the coefficient of determination (R2) and mean absolute error (MAE). The MLP model demonstrated predictive solid power with an R2 score of 88.79% and an average MAE of 7.26, while the 1D-CNN model achieved a slightly higher level of accuracy with an MAE score of 88.35% and an MAE of 6.90. The superior performance of the 1D-CNN model suggests that it captures relevant features from PTA data more effectively than the MLP. These results show that both models hold promise for predicting speech audiometry, potentially simplifying the audiological evaluation process. This approach is applied in clinical settings for hearing loss assessment, the selection of hearing aids, and the development of personalized auditory rehabilitation programs. [ABSTRACT FROM AUTHOR]

An Estimation is a direct summation that can produce new data from past measurements. In this study, we confirm the possibility of using the (WEKA) program in estimating the monthly values of some climatological parameters and investigate the influence of the time series' length parameters on the accuracy of estimation for selected regions of Iraq. Satellite data were used, which represent the monthly values for each of the minimum and maximum temperature, wind velocity, and relative humidity for the 1981 – 2021 time period, the absolute error rate (MAE), and the square root of the error rate (RMSE) with the correlation ( R² ) are also identified to test the confidence of the prediction. Using (WEKA) software, which depends only on the time series of the parameters in the estimation, 12 months were estimated for the mentioned parameters. The estimated values by WEKA were close to the satellite data, thus we can depend on the software as a good source of meteorological data. Also, the study indicates that the time series' length strongly affects the accuracy of the estimation, as the increase in the time series of weather parameters increases the estimate's accuracy, and this increase fluctuates among parameters and varies from one region to another. [ABSTRACT FROM AUTHOR]

In underwater communications for 6G, Doppler effects cause the coherent time to become similar to or shorter than the orthogonal frequency division multiplexing (OFDM) symbol length. Conventional time and frequency synchronization methods require additional training symbols for synchronization, which reduces the traffic data rate. This paper proposes the Zadoff–Chu sequence (ZCS) pilot-based OFDM for time and frequency synchronization. The proposed method transmits ZCS as a pilot for OFDM symbols and simultaneously transmits traffic data to increase the traffic data rate while estimating the CFO at each coherence time. For time–frequency synchronization, the correlation of the ZCS pilot is used to perform coarse and fine time and frequency synchronization in two stages. Since the traffic data cause interference with the correlation of ZCS pilots, we theoretically analyzed the relationship between the amount of traffic data and interference and verified it through computer simulations. The synchronization and BER performance of the proposed ZCS pilot-based OFDM were evaluated by conduction computer simulations and a practical ocean experiment. Compared to the methods of Ren, Yang, and Avrashi, the proposed method demonstrated a 6.3% to 14.3% increase in traffic data rate with similar BER performance and a 2 dB to 3.8 dB SNR gain for a 14.3% to 23.8% decrease in traffic data rate. [ABSTRACT FROM AUTHOR]

The primary objective of this manuscript is to find the service quality factors, which influence customer satisfaction, promotional efforts taken by the retailer, and the perception of customers towards the organized retail sector. This investigation uses both offline and online retail sales datasets. Particularly, in the case of the offline retail sale dataset, an energy‐efficient analysis is conducted to collect scalable, accurate, and real‐time data. The collected data is used in enhancing several aspects of e‐commerce operations, including customer experience optimization and inventory management, which results in better decision‐making with increased efficiency. After data collection, the outliers are eliminated by implementing the z‐score technique. The removal of outliers increases the variability of collected data which reduces statistical power. The statistically sufficient data are given to the Randomized Grasshopper Optimization Algorithm (RGOA) for optimal instance selection. Finally, the selected optimal instances are given to the deep neural network (DNN) model for future customer behaviour prediction. The efficacy of the RGOA‐DNN model is analysed by using evaluation measures like mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE), mean percentage error (MPE) and mean absolute percentage error (MAPE). The numerical analysis states that the RGOA‐DNN model obtained a minimal error rate with MSE of 0.09 and 0.10, RMSE of 0.13 and 0.14, MAE of 0.12 and 0.16, MPE of 0.10 and 0.11 and MAPE of 0.08 and 0.10 on the offline and online retail sales datasets. The RGOA‐DNN model has a minimal error rate in future customer behaviour prediction related to the conventional regression models. Furthermore, the elimination of inactive instances or selection of optimal instances reduces the model complexity to linear and computational time to 22.10 and 23.12 s on the offline and online retail sales datasets. [ABSTRACT FROM AUTHOR]

Density Functional Theory (DFT) is extensively used in theoretical and computational chemistry to study molecular and crystal properties across diverse fields, including quantum chemistry, materials physics, catalysis, biochemistry, and surface science. Despite advances in DFT hardware and software for optimized geometries, achieving consensus in molecular structure comparisons with experimental counterparts remains a challenge. This difficulty is exacerbated by the lack of automated bond length comparison tools, resulting in labor-intensive and error-prone manual processes. To address these challenges, we propose MolGC, a Molecular Geometry Comparator algorithm that automates the comparison of optimized geometries from different theoretical levels. MolGC calculates the mean absolute error (MAE) of bond lengths by integrating data from various DFT software. It provides interactive and customizable visualization of geometries, enabling users to explore different views for enhanced analysis. In addition, it saves MAE computations for further analysis and offers a comprehensive statistical summary of the results. MolGC effectively addresses complex graph labeling challenges, ensuring accurate identification and categorization of bonds in diverse chemical structures. It achieves a 98.91% average rate in correct bond label assignments on an antibiotics dataset, showcasing its effectiveness for comparing molecular bond lengths across geometries of varying complexity and size. The executable file and software resources for running MolGC can be downloaded from https://github.com/AbimaelGP/MolGC/tree/main. [ABSTRACT FROM AUTHOR]

In recent years, vector-to-vector mapping-based raw waveform biosignal enhancement methods have gained significant attention in remote health monitoring system. In this paper, a novel end-to-end convolutional encoder–decoder (CED) model with stacked gated recurrent unit (SGRU) is proposed to learn sequential information of biosignal for signal enhancement. The proposed model CED-SGRU employs convolutional neural network to capture the spatial features and SGRU to capture temporal distributions of the biosignal layer by layer which increases the robustness of the proposed model. This work applies mean absolute error as a loss function for CED-SGRU-based vector-to-vector model. The proposed method has been evaluated on three foremost required biosignals, namely electrocardiogram, photoplethysmography and heart rate signals for diagnosing cardiovascular diseases. Experimental result shows outstanding denoising capability of the proposed CED-SGRU model on three biosignals which yields significantly higher reconstruction signal-to-noise ratio value and lower average absolute error, root mean square error and percent root mean square difference values when compared with state-of-the-art-methods. Moreover, the simple architecture of SGRU lowers the complexity of the model; thereby, reducing the inference time for denoising and restoring compressed biosignal tasks is fairly compared with baseline models, namely recurrent neural network model and long short-term memory model. [ABSTRACT FROM AUTHOR]

This exploration of paper presents an investigation of the Forest Region Inclusion Dataset that gives data on the backwoods inclusion of different nations overall from 1990 to 2020. The dataset contains country-wise information on population, population density, population development rate, total population rate, and forest region inclusion. We examined this dataset to decide the patterns in woodland region inclusion across various nations and mainlands, as well as the connection among populace and backwoods region inclusion. Our discoveries show that while certain nations have essentially expanded their forest region inclusion, others have encountered a decline. Besides, we found that population density and development rate are adversely related with forest area coverage. Authors have implemented four machine learning algorithms that are Linear Regression, Decision Tree, Random Forest and Support Vector Machine on the dataset. [ABSTRACT FROM AUTHOR]

The crucial importance of agriculture for survival is complemented by the transformational capacity of machine learning (ML) in addressing agricultural yield concerns. Conventional techniques, such as manual enumeration and satellite imagery, often lack precision. This article centres on the use of machine learning (ML) to predict millet output. This allows farmers to enhance their crop yields by considering parameters such as land area and irrigation. Millets, known for their nutritional robustness, provide a substantial contribution to global food security. Precise yield forecasts are crucial for the long-term viability of agriculture. The research utilises regression models like AdaBoost Regressor, XGBoost Regressor, Decision Tree, Support Vector Machine and Random Forest. Among these models, AdaBoost Regressor demonstrates the best level of accuracy. The combination of machine learning and modern technologies improves the accuracy of yield estimate and highlights the connection between agricultural practices and state-of-the-art technology. [ABSTRACT FROM AUTHOR]

This article is dedicated to the study of traffic flow forecasting at the intersection of Bogishamol Street in Tashkent, Uzbekistan. In the context of the rapid growth of vehicular traffic and frequent congestion, developing effective forecasting models is a pressing task that will help optimize traffic flow management. The research examines and analyzes various machine learning methods, such as decision trees, random forests, and gradient boosting, for predicting traffic intensity. The data for the models was collected using video cameras installed at the intersection which provided accurate and up-to-date traffic flow information. The main focus of the study is on the comparative analysis of the performance of these methods. The comparison was made using various evaluation metrics, such as the coefficient of determination (R2), mean squared error (MSE), and mean absolute error (MAE). These metrics allowed for an objective assessment of the accuracy and effectiveness of each method in the context of traffic flow prediction. The results of the study showed that the gradient boosting model demonstrated the best performance among the methods considered. It achieved the highest R2 values and the lowest MSE and MAE values, indicating its high accuracy and ability to adequately predict changes in traffic flows. The decision tree and random forest models also showed good results but were outperformed by gradient boosting in key indicators. The findings have significant practical implications. They can be used to develop intelligent traffic management systems aimed at increasing the capacity of roads and intersections. This, in turn, can help reduce congestion, lower emissions of harmful substances into the atmosphere, and decrease economic costs associated with traffic delays. [ABSTRACT FROM AUTHOR]

Abstract This study investigates the impact of spatio- temporal correlation using four spatio-temporal models: Spatio-Temporal Poisson Linear Trend Model (SPLTM), Poisson Temporal Model (TMS), Spatio-Temporal Poisson Anova Model (SPAM), and Spatio-Temporal Poisson Separable Model (STSM) concerning food security and nutrition in Africa. Evaluating model goodness of fit using the Watanabe Akaike Information Criterion (WAIC) and assessing bias through root mean square error and mean absolute error values revealed a consistent monotonic pattern. SPLTM consistently demonstrates a propensity for overestimating food security, while TMS exhibits a diverse bias profile, shifting between overestimation and underestimation based on varying correlation settings. SPAM emerges as a beacon of reliability, showcasing minimal bias and WAIC across diverse scenarios, while STSM consistently underestimates food security, particularly in regions marked by low to moderate spatio-temporal correlation. SPAM consistently outperforms other models, making it a top choice for modeling food security and nutrition dynamics in Africa. This research highlights the impact of spatial and temporal correlations on food security and nutrition patterns and provides guidance for model selection and refinement. Researchers are encouraged to meticulously evaluate the biases and goodness of fit characteristics of models, ensuring their alignment with the specific attributes of their data and research goals. This knowledge empowers researchers to select models that offer reliability and consistency, enhancing the applicability of their findings.

Facial age recognition has been widely used in real-world applications. Most of current facial age recognition methods use deep learning to extract facial features to identify age. However, due to the high dimension features of faces, deep learning methods might extract a lot of redundant features, which is not beneficial for facial age recognition. To improve facial age recognition effectively, this paper proposed the deep manifold learning (DML), a combination of deep learning and manifold learning. In DML, deep learning was used to extract high-dimensional facial features, and manifold learning selected age-related features from these high-dimensional facial features for facial age recognition. Finally, we validated the DML on Multivariate Observations of Reactions and Physical Health (MORPH) and Face and Gesture Recognition Network (FG-NET) datasets. The results indicated that the mean absolute error (MAE) of MORPH is 1.60 and that of FG-NET is 2.48. Moreover, compared with the state of the art facial age recognition methods, the accuracy of DML has been greatly improved.

This study investigates the impact of spatio- temporal correlation using four spatio-temporal models: Spatio-Temporal Poisson Linear Trend Model (SPLTM), Poisson Temporal Model (TMS), Spatio-Temporal Poisson Anova Model (SPAM), and Spatio-Temporal Poisson Separable Model (STSM) concerning food security and nutrition in Africa. Evaluating model goodness of fit using the Watanabe Akaike Information Criterion (WAIC) and assessing bias through root mean square error and mean absolute error values revealed a consistent monotonic pattern. SPLTM consistently demonstrates a propensity for overestimating food security, while TMS exhibits a diverse bias profile, shifting between overestimation and underestimation based on varying correlation settings. SPAM emerges as a beacon of reliability, showcasing minimal bias and WAIC across diverse scenarios, while STSM consistently underestimates food security, particularly in regions marked by low to moderate spatio-temporal correlation. SPAM consistently outperforms other models, making it a top choice for modeling food security and nutrition dynamics in Africa. This research highlights the impact of spatial and temporal correlations on food security and nutrition patterns and provides guidance for model selection and refinement. Researchers are encouraged to meticulously evaluate the biases and goodness of fit characteristics of models, ensuring their alignment with the specific attributes of their data and research goals. This knowledge empowers researchers to select models that offer reliability and consistency, enhancing the applicability of their findings. [ABSTRACT FROM AUTHOR]

Effective machine learning regression models are useful toolsets for managing and planning energy in PV grid-connected systems. Machine learning regression models, however, have been crucial in the analysis, forecasting, and prediction of numerous parameters that support the efficient management of the production and distribution of green energy. This article proposes multiple regression models for power prediction using the Sharda University PV dataset (2022 Edition). The proposed regression model is inspired by a unique data pre-processing technique for forecasting PV power generation. Performance metrics, namely mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), R2-score, and predicted vs. actual value plots, have been used to compare the performance of the different regression. Simulation results show that the multilayer perceptron regressor outperforms the other algorithms, with an RMSE of 17.870 and an R2 score of 0.9377. Feature importance analysis has been performed to determine the most significant features that influence PV power generation. [ABSTRACT FROM AUTHOR]

This paper describes modeling solar power generation as a renewable energy generator by simulating the analytical approach mean absolute error and root mean square error (MAE and RMSE). This research estimates the error referring to long short-term memory (LSTM) network learning. Related to this, the Indonesian government is currently actively developing solar power plants without ignoring the surrounding environment. The integration of solar power sources without accurate power prediction can hinder the work of the grid and the use of new and renewable generation sources. To overcome this, virtual power plant modeling can be a solution to minimize prediction errors. This study proposes a method for on-site virtual solar power plant efficiency with a research approach using two models, namely RMSE and MAE to account for prediction uncertainty from additional information on power plants using virtual solar power plants. A prediction strategy verified against the output power of photovoltaic (PV) modules and a set based on data from meteorological stations used to simulate the virtual power plants (VPP) model. This forecast prediction refers to the LSTM network and provides forecast errors with other learning methods, where the approach simulated with 12.36% and 11.85% accuracy for MAE and RMSE, respectively. [ABSTRACT FROM AUTHOR]

This paper formulated a support vector machine powered time-aware food recommender system (SVMTAFRS) to recommend healthy food for the customers. The rated food item incorporates the user preference (UP) in terms of calories, nutrition factor, and all food contents required for a healthy diet. This also takes into account the user age, time of day and week day while predicting the food rating. The SVMTAFRS involves two steps for computation of user identity document (UID) and predicted food rating (PFR). UID is computed considering the customer age (CA), UP in terms of calories and suitable weight factors. PFR is computed considering the UID and time of day (TOD). PFR for week end day is computed by multiplying the PFR by week end multiplying factor (WEMF). Support vector machine (SVM) is used for recommending the suitable healthy food for customer in terms of correct values of PFR. Efficacy of PFR is tested in terms of mean absolute error (MAE) and root mean squared error (RMSE). This is established that performance of the SVMTAFRS is superior compared to the rule-based food recommender system (RBFRS). [ABSTRACT FROM AUTHOR]

Accurate reporting and prediction of PM2.5 concentration is very important for improving public health. In this article, we use spectral clustering algorithm to cluster 15 cities in the Pearl River Delta. On this basis, we propose a special difference equation model, especially the use of nonlinear diffusion equations to characterize the temporal and spatial dynamic characteristics of PM2.5 propagation between and within clusters for real‐time prediction. For example, through the analysis of PM2.5 concentration data for 91 consecutive days in the Pearl River Delta, and according to different accuracy definitions, the average prediction accuracy of the difference equation model in all city clusters is 97% or 88%. The mean absolute error (MAE) of the forecast data for each urban agglomeration is within 7 units (μg/m3). Experimental results show that the difference equation model can effectively reduce the prediction time and improve the prediction accuracy. Therefore, based on the spectral clustering algorithm and the difference equation model, the fastest prediction speed and the best prediction result can be obtained, and the problem of PM2.5 concentration prediction can be effectively solved. The research can provide decision support for local air pollution early warning and urban integrated management. [ABSTRACT FROM AUTHOR]

AbstractThis paper presents new simple lower and upper bounds for the cumulative normal distribution function, Φ(z). The accuracy and closeness of the proposed bounds to the exact Φ(z) are investigated based on the maximum absolute error and the mean absolute error. It is found that the maximum absolute error of the proposed lower bound is 8.55×10−3 and it is 4.1×10−4 for the upper bound. In addition, based on 5001 values between z=0 and z=5 with step 0.001, we found that the mean absolute error is 3.27×10−3 for the lower bound and it is 1.1×10−4 for the upper bound and these two values decrease with increasing the z value. [ABSTRACT FROM AUTHOR]

Copyright of Tecnología y Ciencias del Agua is the property of Instituto Mexicano de Tecnologia del Agua (IMTA) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Assessing hair count and direction in the transplant area is crucial for surgery success. Currently, this requires intensive manual work by medical staff. Utilizing a deep learning object detection model for automatic hair detection and counting could greatly enhance clinical efficiency. In this study, we analyzed hair detection effectiveness by comparing two labeling techniques: the conventional Horizontal Bounding Box (HBB) and the direction-sensitive Oriented Bounding Box (OBB). We assessed the performance of models based on these methods using metrics such as mean Average Precision (mAP) with an Intersection Over Union (IOU) threshold, and Mean Absolute Error (MAE), to determine the most efficient approach for hair detection. Results showed HBB achieved an mAP of 0.468 and MAE of 2.89, while OBB recorded an mAP of 0.478 and MAE of 3.30. This paper explores the pros and cons of HBB and OBB for hair detection, proposing their use to improve automation efficiency in clinical settings. [ABSTRACT FROM AUTHOR]

Cryptocurrency is a form of digital currency using cryptographic techniques in a decentralized system for secure peer-to-peer transactions. It is gaining much popularity over traditional methods of payment because it facilitates very fast, easy, and secure transactions. Social media is a significant influence, but it is also very volatile and subject to a variety of other factors. Thus, with over four billion active users on social media, we need to understand its influence on the crypto market and how it can lead to fluctuations in the values of these cryptocurrencies. In our work, we analyze the influence of activities on Twitter, in particular the sentiments of the tweets posted regarding cryptocurrencies and how they influence their prices. In addition, we also collect metadata related to tweets and users. We try to leverage these features to predict the price of cryptocurrency, for which we use some regression-based models and an LSTM-based model. [ABSTRACT FROM AUTHOR]

Solar energy represents one of the emerging frontiers in renewable energy, offering significant potential to address the issues of energy unavailability and instability in Uyo (Nigeria). A crucial step in overcoming these challenges is accurately predicting the amount of solar energy that can be harnessed at a specific location. This research focused on achieving optimal solar power prediction, with the following objectives; identifying and investigating the mathematical relationships between relevant variables and parameters. To ensure precise predictions, artificial neural networks (ANN) were employed, utilizing both forward and backward propagation techniques. The input data for the ANN comprised radiation data obtained from a secondary source, the solar panel's size or area from the manufacturer, the panel's efficiency, and its performance ratio – all of which determined the electricity produced in kilowatts. The ANN was trained and tested using meteorological data, enabling accurate predictions of optimal electricity generation for the location. Notably, the hourly predictions reached their peak by 1 PM at the geographic location (5.2N and 7.5E), indicating that the highest levels of solar power were attainable during this daily period. Moreover, the pattern of monthly average solar power exhibited optimal predictions in January. Influenced by meteorological factors, a significant rise and fall in August, commonly referred to as the 'August Break’ featured. The results demonstrated exceptional accuracy with minimal error margins (mean absolute error (MAE) of 0.03, mean squared error (MSE) of 0.0, and root mean squared error (RMSE) of 0.03). This high level of accuracy rendered the predictions reliable, making them suitable for consultancy services. Additionally, the potential for future work and expansion was evident, as the ANN could incorporate five or more years of radiation data for further improvements and insights. [ABSTRACT FROM AUTHOR]

Since the introduction of the BrainAGE method, novel machine learning methods for brain age prediction have continued to emerge. The idea of estimating the chronological age from magnetic resonance images proved to be an interesting field of research due to the relative simplicity of its interpretation and its potential use as a biomarker of brain health. We revised our previous BrainAGE approach, originally utilising relevance vector regression (RVR), and substituted it with Gaussian process regression (GPR), which enables more stable processing of larger datasets, such as the UK Biobank (UKB). In addition, we extended the global BrainAGE approach to regional BrainAGE, providing spatially specific scores for five brain lobes per hemisphere. We tested the performance of the new algorithms under several different conditions and investigated their validity on the ADNI and schizophrenia samples, as well as on a synthetic dataset of neocortical thinning. The results show an improved performance of the reframed global model on the UKB sample with a mean absolute error (MAE) of less than 2 years and a significant difference in BrainAGE between healthy participants and patients with Alzheimer's disease and schizophrenia. Moreover, the workings of the algorithm show meaningful effects for a simulated neocortical atrophy dataset. The regional BrainAGE model performed well on two clinical samples, showing disease‐specific patterns for different levels of impairment. The results demonstrate that the new improved algorithms provide reliable and valid brain age estimations. [ABSTRACT FROM AUTHOR]

The Power Function distribution is one of the most suitable distributions for failure times modeling, especially of electric components and devices. The article proposes some new modified estimators for parameter estimation of the Power Function distribution. The proposed modified estimators are based on the mean and geometric mean of the distribution. The performance of these estimators is compared with existing traditional and modified estimators by means of Monte Carlo simulation and two real‐life examples. Total mean square error, total relative deviation, root mean square error and mean absolute error are used as performance criteria. The performance of the estimators is also compared on two real‐life data sets repressing the device failures times. From both, Monte Carlo simulation and real‐life applications, the results indicated that the proposed modified estimators perform better than the competing estimators and hence their use is recommended for parameter estimation of the Power Function probability model. The findings will help to get more precise parameter estimates for the distribution which is very much applied in different fields particularly in modelling failure times of electronic devices and components. [ABSTRACT FROM AUTHOR]

PurposeTo assess whether the predictive accuracy of machine learning algorithms using Kalman filtering for forecasting future values of global indices on perimetry can be enhanced by adding global retinal nerve fiber layer (RNFL) data and whether model performance is influenced by the racial composition of the training and testing sets.DesignRetrospective, longitudinal cohort study.ParticipantsPatients with open-angle glaucoma (OAG) or glaucoma suspects enrolled in the African Descent and Glaucoma Evaluation Study or Diagnostic Innovation in Glaucoma Study.MethodsWe developed a Kalman filter (KF) with tonometry and perimetry data (KF-TP) and another KF with tonometry, perimetry, and global RNFL data (KF-TPO), comparing these models with one another and with 2 linear regression (LR) models for predicting mean deviation (MD) and pattern standard deviation values 36 months into the future for patients with OAG and glaucoma suspects. We also compared KF model performance when trained on individuals of European and African descent and tested on patients of the same versus the other race.Main outcome measuresPredictive accuracy (percentage of MD values forecasted within the 95% repeatability interval) differences among the models.ResultsAmong 362 eligible patients, the mean ± standard deviation age at baseline was 71.3 ± 10.4 years; 196 patients (54.1%) were women; 202 patients (55.8%) were of European descent, and 139 (38.4%) were of African descent. Among patients with OAG (n = 296), the predictive accuracy for 36 months in the future was higher for the KF models (73.5% for KF-TP, 71.2% for KF-TPO) than for the LR models (57.5%, 58.0%). Predictive accuracy did not differ significantly between KF-TP and KF-TPO (P = 0.20). If the races of the training and testing set patients were aligned (versus nonaligned), the mean absolute prediction error of future MD improved 0.39 dB for KF-TP and 0.48 dB for KF-TPO.ConclusionsAdding global RNFL data to existing KFs minimally improved their predictive accuracy. Although KFs attained better predictive accuracy when the races of the training and testing sets were aligned, these improvements were modest. These findings will help to guide implementation of KFs in clinical practice.

In today's digital entertainment environment, movie recommendation algorithms are essential for intensifying user ecstasy and content engagement. Online streaming services like Netflix can potentially increase revenue by implementing a recommendation system that offers tailored movie suggestions to users based on their past interactions with the platform. The strategies we'll discuss here are not simply limited to films; they may be used for any product for which you want to develop a Recommendation System (RS). This paper gives a preamble of several types of recommendation strategies based on user preferences, ratings, domain knowledge, users’ demographic data, user’s context. This work also envisions the singular value decomposition plus-plus (SVD++) and collaborative filtering-based movie recommendation method. The suggested method is compared to well-known machine learning methods such as k closest neighbour (K-NN), singular value decomposition (SVD), and co-clustering. The suggested method is experimentally tested using MovieLens 100 K datasets, and the error of the RS is assessed using Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). In order to address issues like scalability, data sparsity, and the cold start problem, our study is driven by the requirement to determine the most efficient method for offering individualized movie suggestions. This technique is used by numerous e-commerce companies, including Amazon, Flipkart, and Mantra, to understand their customers' purchasing patterns and provide recommendations to them about the goods they are most likely to purchase. [ABSTRACT FROM AUTHOR]

In this research, the reliability functions are applied to estimate the hazard function of four used car components such as (tires, brakes, lights, and engine), which are inspected by aperiodic vehicle inspection (PVI) established in Erbil city, a specialized company that conducts the annual technical inspection of vehicles to detect the failure component, that either require repair or replace it with a new one. For our purpose, the data about the failure components of a sample of size (50,000) cars are obtained from the Erbil traffic directorate, which are annually inspected for 11 years (2010–2020) by a (PVI) company. From the available data, the reliability function, hazard function, and probability density function of the failure time of each component are found by the non-parametric method and the estimated Rayleigh distribution since the failure rates of the components are the linear functions of time, also the comparison between their reliability values have made by the mean absolute error method. [ABSTRACT FROM AUTHOR]

Electrified transportation systems are emerging quickly worldwide, helping to diminish carbon gas emissions and paving the way for the reduction of global warming possessions. Battery remaining useful life (RUL) prediction is gaining attention in real world applications to tone down maintenance expenses and improve system reliability and efficiency. RUL forms the prominent component of fault analysis forecast and health management when the equipment operation life cycle is considered. The uprightness of RUL prediction is vital in providing the effectiveness of electric batteries and reducing the chance of battery illness. In assessing battery performance, the existing prediction approaches are unsatisfactory even though the battery operational parameters are well tabulated. In addition, battery management has an important contribution to several sustainable development goals, such as Clean and Affordable Energy (SDG 7), and Climate Action (SDG 13). The current work attempts to increase the prediction accuracy and robustness with selected machine learning algorithms. A Real battery life cycle data set from the Hawaii National Energy Institute (HNEI) is used to evaluate accuracy estimation using selected machine learning algorithms and is validated in Google Co-laboratory using Python. Evaluated error metrics such as Mean Square Error (MSE), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), R-Squared, and execution time are computed for different L methods and relevant inferences are presented which highlight the potential of battery RUL prediction close to the most accurate values. [ABSTRACT FROM AUTHOR]