Search

Media coverage plays an important role in prevention and control the spread of COVID-19 during the pandemic. In this paper, an SIHRS model of COVID-19 pandemic with impulse and time delay under media coverage is established. The positive and negative emotions of public are considered by the impact of confirmed cases and medical resources. In order to restrain the negative information of public, the factor of policies and regulations with impulse and time delay is introduced. Furthermore, the system model is simulated and verified by the reported data of COVID-19 pandemic in Wuhan. The main results are as follows: (1) When the implementation rate of the negative information generated by the confirmed cases gradually reduced to 0.4 times, the cumulative confirmed cases will be significantly reduced to about 37000, indicating that the popularization of pandemic related media information should be broad; (2) When the implementation rate affected by the amount of policies and regulations information gradually increases to 3 times, the cumulative confirmed cases will be significantly reduced to about 28000, indicating that the policies and regulations information should be continuously and incrementally reported; (3) When the inhibition rate of policies and regulation information on negative information gradually increases to 3 times, the cumulative confirmed cases will also be significantly reduced to about 27000 cases, indicating that the targeted policies and regulations information has a significant impact on inhibiting the corresponding negative emotions.

The decrease of situation awareness (SA) is one of reasons leading to human factor accidents in nuclear power plants. The main purpose of this paper is to the evaluation and prediction the operators’ SA in digital main control room. Firstly, this paper used both the entropy weight method and variation coefficient method to determine the relevant influencing factors. Secondly, principal component analysis (PCA) was used to concentrate the input variables into common component. Then, an artificial neural network (ANN) model was conducted based on influencing factors and SA data. The result showed that there are identified fifteen factors that have a greater impact on SA reliability, accounting for 65.2% of the weight of all factors. The PCA result showed that the contribution rate of eight common factors reached 80.6% for the total variance of variables and the cumulative variance. Therefore, these variables were explained by eight common components. The 8-14-1 network structure was can obtain the minimum of the MSE (0.0058) and the maximum of R2 (0.9814). The predicted data can obtain the minimal MSE value (0.0035) and maximum R2 (0.9886) when the ‘Relu’ function was used as the activation function of both the hidden layer and output layer. The average prediction accuracy of the constructed ANN model is more than exceeded 92% for the test data. This result indicated that the developed ANN model can accurately evaluate operator’s SA.

To assess the quality of a binary classification, researchers often take advantage of a four-entry contingency table called confusion matrix, containing true positives, true negatives, false positives, and false negatives. To recap the four values of a confusion matrix in a unique score, researchers and statisticians have developed several rates and metrics. In the past, several scientific studies already showed why the Matthews correlation coefficient (MCC) is more informative and trustworthy than confusion-entropy error, accuracy, F1 score, bookmaker informedness, markedness, and balanced accuracy. In this study, we compare the MCC with the diagnostic odds ratio (DOR), a statistical rate employed sometimes in biomedical sciences. After examining the properties of the MCC and of the DOR, we describe the relationships between them, by also taking advantage of an innovative geometrical plot called confusion tetrahedron, presented here for the first time. We then report some use cases where the MCC and the DOR produce discordant outcomes, and explain why the Matthews correlation coefficient is more informative and reliable between the two. Our results can have a strong impact in computer science and statistics, because they clearly explain why the trustworthiness of the information provided by the Matthews correlation coefficient is higher than the one generated by the diagnostic odds ratio.

The Mixed distribution model of time headway plays an important role in Intelligent Transportation System. A more accurate improved model based on a new traffic state classification method and correlation coefficient method was established in this study. Upon analyzing traffic data from three unban roads in Nanjing, China from December 8, 2017 to April 24, 2018 in clear days free of fogs and haze, it was found that 1) the improved model dwarfs the other models, for it boasts higher goodness-of-fit and it represents the only one to successfully pass the Chi-square test. 2) the improved model shows better performance in application because the relative errors between the field traffic rate and calculated traffic rate obtained from the improved model are the smallest. 3) the goodness-of-fit of the improved model is affected by traffic state, and the fitting precision of the improved model under four states is the best. Thus, it is hard to tell performance of which model established under two states or three-state model is better. 4) correlation coefficients between time headway and the absolute value of relative speed make the improved Mixed model more fitted and precise. The improved model proposed in this study is a more accurate time headway distribution model.

Pattern matching is a fundamental problem in theoretical computer science. The algorithms for pattern matching and the study on the statistics of pattern matching have found enormous applications in practical fields. In this paper, we revisit the Markov embedding approach for studying pattern matching in repeated experiments. For any pattern of length $m$ over alphabet $\Sigma $ , we show that the mean and variance of the waiting time of the pattern in iid experiments can be computed in $O(m)$ time based on Markov embedding technique, improving over the $O(|\Sigma | \cdot m)$ and $O(m^{2})$ naïve bounds. Our method extends to computing the $k$ -th moment of the waiting time, and it extends to computing other related statistics about pattern matching in repeated experiments, and it also extends to the case of Markov dependent experiments.

5G millimeter wave (mmWave) application in mobile connectivity to realize high-speed, reliable communication is attributed with high path loss. This paper presents a detailed 3D ray-tracing technique at 28 GHz for Lagos Island to investigate five unique path loss scenarios: path loss, free space path loss with antenna pattern, free space path loss without antenna pattern, excess path loss with antenna pattern, and the excess path loss without antenna pattern for an urban environment. The Close-In (CI) model, Floating Intercept (FI) path loss model, and a root mean square error (RMSE) are used to model and evaluate the best path loss model for Lagos Island. The average achieved FI ( $\alpha,\beta, \sigma $ ) parameters were 189.92352, 0.1654, and 0.66948, While the average CI ( $\eta,X\sigma $ ) parameters were 2.309355 and 56.236425. From all the scenarios evaluated, the lowest path loss exponent achieved was 0.45, while the highest path loss exponent was 3.8. We have established that the FI path loss model accurately characterizes path loss for the Lagos Island environment with the lowest RMSE of 0.0359 dB and the highest RSME of 0.0997 dB. In contrast, the CI model over-predict the path loss at 28 GHz with the lowest RMSE of 0.0495 dB and the highest RMSE of 2.2547 dB. This work opens up a new area of research on mm-Wave at 28 GHz in Lagos Island, and the results obtained from this work can be used to benchmark future studies on mmWave in a similar environment.

This paper investigates the opinion dynamics on the adaptive networks with the community structure. A coevolution model of opinions and networks is proposed, in which the influence weights among agents can not be precisely obtained, and there exist the confidence heterogeneities and the frequency difference of exchanging opinion between the inner-community and the inter-community. A two-dimensional fuzzy inference system is designed to obtain the influence weights among agents’ opinions, which deals with the difficulties that the influence weights can not be precisely obtained. The simulation results show that the imbalance levels between the inner-community confidence and the inter-community confidence not only affect strongly the average relative frequency of survival opinions, but also dominate the distribution of survival opinions. The polarization state of opinions easily appears in the population with the higher inner-community confidence, whereas when the inter-community confidence is higher, the system tends to exhibit a large number of small-size clusters and a plural state is achieved. Moreover, the simulation results also show that the convergence time to the stable state varies non-monotonically with the increase of the inner-community confidence or the inter-community confidence.

In this paper, a two-stage method is proposed for predicting the catch of skipjack tuna (Katsuwonus pelamis) from a 2D sea environmental pattern. Following the assumption that sea water temperature and sea surface height (SSH) which fishermen often use for finding fishing spots has a correlation with the skipjack tuna catch, a new approach of using Faster R-CNN in object detection is proposed. The proposed method consists of two part. In the first part, taking a sea temperature map as input, Faster R-CNN extracts the candidates of where skipjack tuna would be on the map in order to imitate the behaviors of fishers. In the second part, Support Vector Regression (SVR) estimates the catch amount in each candidate. Fater R-CNN is applied to several sea environmental patterns with three different loss functions and compares each performance. The proposed model is evaluated by comparing the result with average fishers’ ability on the skipjack tuna catches and several criteria for evaluating the proposed model. The results show that the proposed method is able to outperform the average fishers’ ability by an average of 3%.

Weed identification in vegetable plantation is more challenging than crop weed identification due to their random plant spacing. So far, little work has been found on identifying weeds in vegetable plantation. Traditional methods of crop weed identification used to be mainly focused on identifying weed directly; however, there is a large variation in weed species. This paper proposes a new method in a contrary way, which combines deep learning and image processing technology. Firstly, a trained CenterNet model was used to detect vegetables and draw bounding boxes around them. Afterwards, the remaining green objects falling out of bounding boxes were considered as weeds. In this way, the model focuses on identifying only the vegetables and thus avoid handling various weed species. Furthermore, this strategy can largely reduce the size of training image dataset as well as the complexity of weed detection, thereby enhancing the weed identification performance and accuracy. To extract weeds from the background, a color index-based segmentation was performed utilizing image processing. The employed color index was determined and evaluated through Genetic Algorithms (GAs) according to Bayesian classification error. During the field test, the trained CenterNet model achieved a precision of 95.6%, a recall of 95.0%, and a $F_{1}$ score of 0.953, respectively. The proposed index -19R + 24G -2B ≥ 862 yields high segmentation quality with a much lower computational cost compared to the wildly used ExG index. These experiment results demonstrate the feasibility of using the proposed method for the ground-based weed identification in vegetable plantation.

Ingredient amounts are crucial for food-oriented health systems, but this information is seldom used in food-oriented health systems due to the difficulty of fetching it from online recipes. This study proposes a predictive model named DeepRecipes to extract ingredient amounts from online textual recipes. The model predicts ingredient amounts according to a given recipe’s name and listed ingredients. We train the model on a small set of recipes containing all ingredients and their corresponding amounts. As we can extract the recipe names and ingredients from almost all online recipes, the proposed model can potentially recover ingredient amounts for massive online recipes. We first trained the model on a small set of recipes containing all ingredients and their corresponding amounts. Then, we compared ten models as references for their performances. The performance of DeepRecipes exceeds those of all the comparison models. The model’s mean absolute error (MAE) and mean absolute percentage error (MAPE) are $3.96\times {10}^{-1}$ and 18.57%, respectively, and its APEs are lower than 50% in more than 95% of the total predictions. This accuracy is sufficient for providing rough ingredient amount estimations for food-oriented health systems.

Background: Accurate and reliable prediction of changes in the severity of cerebellar ataxia (CA) will be necessary for trials of disease-modifying therapies. Cerebellar dysarthria (CD) is a common feature of CA. This study demonstrated that objective acoustic measures were more sensitive than perceptive analysis through the Scale for the Assessment and Rating of Ataxia (SARA) in assessing the progression of CD, within a time window of two years (mean). Method: Thirty-seven people with CA were tested at baseline (time point 1, TP1) and two years later (time point 2, TP2). A machine-learning framework with a robust three-step feature selection criterion and a Bayesian data-driven clustering technique based on the multivariate mixture extension of the generalized linear mixed model (GLMM) was used. The outcomes included two (time and cepstral-based) objective speech parameters recorded at TP1 and TP2. Subject testing involved dynamic prediction and was conducted using samples from the posterior distributions of parameter estimates and random effects. This study further employed the penalized expected deviance (PED) criterion for model comparison and the selection of the number of groups in the clustering procedure. Results: First, the selected objective speech metrics in the individual patients showed a significant worsening of the speech impairment (p $PED_{Fea1+Fea2}=5175$ ; number of groups = 2) performed more reliably than the univariate models ( $PED_{Fea1}=4225$ , $PED_{Fea2}=3850$ ; number of groups = 2) in discriminating the CA patients. Fourth, the individual-level predictions of the change in profiles of the objective measures over time were performed for the testing data. Conclusion: Such a framework using objective speech metrics indeed holds promise to predict the rate of clinical progression of CD in individuals with CA.

The article discusses a wavelet-based approach for recognizing sugarcane stem nodes in order to improve pre-cut sugarcane planting technology, beginning with sugarcane form characteristics that permit automated sugarcane seed production. The location signal is collected by the acceleration and thin-film piezoelectric sensors and then decomposed into the tenth, eleventh, and twelfth layers using the Daubechies tight-branch wavelet. After capturing the signal, it is reconstructed and superimposed to capture the stem node region’s features using the default threshold technique. A multi-sensor fusion approach is developed based on a weighted average and a Kalman filter to confirm the experiment’s validity. The weighted average process produces an average value that is 0.3512 mm off from the experimentally observed data average. The discrepancy between the Kalman filter method’s anticipated average value and the empirically determined average error is 0.5778 mm. To facilitate the investigation, 175 sugarcane samples with intermediate length processing are used. The detecting position system is determined experimentally after extensive experimental research and diligent examination. On average, the standard deviation is 0.494 mm, while the maximum value is 9.99 mm. 99.63 percent of cane seed samples are detected, with an error rate of 0.37 percent and a response time of 0.25 seconds. The proposed technology is conceptually feasible and achievable, and it can provide a reference for the development of automated cane seed pre-cutting machinery to give its contribution to agricultural production.

Ensuring site diversity is a fade mitigation technique used to overcome severe rain-induced attenuation encountered in high-frequency signal communications, especially over tropical regions. However, due to the high cost of deploying diverse equipment and terrestrial infrastructures for connections between two sites, gain prediction models are widely used to evaluate the performance of a given scheme. This article presents a study on the behavior of four parameters that contribute to site diversity gain: site separation distance, link frequency, elevation angle, and baseline orientation angle. A correlation between gain and distance is found in the form of an attenuation-dependent logarithmic function instead of the gain being saturated when the increment of the distance exceeds the convective rain cell extent, as was found in the literature. Therefore, a site diversity logarithmic model, SDLog, is proposed. The analysis utilized seven site diversity experiments conducted in the tropics, i.e., in Malaysia, Singapore, and Guam, USA. The performance of the SDLog model was compared with the performances of existing models, namely, the ITU-R, Hodge, Panagopoulos, Semire, and X. Yeo models. The SDLog model reproduces the experimental datasets with an average root mean squared error (RMSE) of 0.12, while those of the ITU-R, Hodge, Panagopoulos, Semire, and X. Yeo models are 0.226, 0.289, 0.19, 0.192, and 0.311, respectively. These models were also evaluated based on diversity experiments conducted in Indonesia, separate from the former evaluation. The RMSE and mean absolute percentage error (MAPE) of all evaluations are presented, and the SDLog model seems to be convincing.

Ground vibrations caused by blasting operations in cement canisters is among the main mining issues that cause significant disruptions to nearby buildings and infrastructure. This research was performed in a limestone quarry situated southeast of Helwan City, Egypt, to investigate the impact of ground motion vibration due to cement blast action in limestone rocks. To reduce the environmental impact of quarry blasting, continuous monitoring, and accurate medium’s peak particle velocity ( PPV ) assessment are required. Recently, machine learning ( ML ) models are employed in diverse applications. The default hyperparameters of such models must be modified to fit the problem concerned. The hyperparameters optimization for ML models impacts the employed model’s performance and efficiency. In this research, different regression models are implemented for predicting the PPV values. A dataset representing 1438 blast incidents in the Helwan area was built and utilized to evaluate the considered ML models. This dataset incorporates the relationship of the ground vibration amplitude to both the explosive charge weight per delay and distance from the blast. The predictive models’ output performance has been evaluated using the root-mean-squared error ( RMSE ) and the coefficient of determination ( $R^{2}$ ). The PPV dataset has been divided into training and testing data to produce statistically significant results and to make the dataset more representative to avoid overfitting. The utilized test PPV dataset acts as a proxy for any new PPV data prediction. There was evidence of higher performance in the developed Decision Trees model with the lowest RMSE and the highest $R^{2}$ on training and testing data. The decision tree is, therefore, an acceptable algorithm for the construction of a predictive PPV model for other quarry blasting areas with conditions identical to those in Helwan. Finally, comparative experimental results have shown that optimized models can predict PPV values with lower errors and greater prediction accuracy.

The objective of this research work is to analyze wind characteristics and to assess wind power potential by selecting the best fit probability distribution function of Jhimpir Sindh Pakistan. This type of detailed investigation helps wind power generation companies in selecting suitable wind turbine and provides information of wind characteristics of potential site. Eight probability distribution functions are tested on the wind speed data from January 2015 to July 2018. Frequency bins of Weibull and Rayleigh distribution with maximum probabilities of 0.1210 and 0.1143 are most closest representation of our data. In order to, quantitatively analysis which distribution function is best fitting the local wind regime, we have applied the coefficient-of-determination, Kolmogorov-Smirnov, Chi square, Cramer-von Mises, Anderson-Darling tests along with Akaike information and Bayesian information criterion. These statistical test are used to rank the empirical distribution functions in order to identify two distribution function better fitting the actual wind speed data. After selecting two best fitted distribution functions, we analyze wind power potential and compare the error of wind power density based on these distribution functions (Weibull and Rayleigh). The power densities reported varied from 73.67 to 648.73W/m2. Results indicate that power densities of Weibull and Rayleigh for the candidate site are 84.67–698.65W/m2 and 83.67–1021.4W/m2, respectively. The highest error for Weibull and Rayleigh are 0.1850 and 0.5745, respectively. Whereas lowest error are 0.0178 and 0.0180, respectively. Complete analysis suggested that Weibull distribution function is the most suitable for Jhimpir Sindh Pakistan and the studied site is suitable for wind power production. In addition, comprehensive analysis of wind direction at the candidate site suggested that Eastern and Southeastern wind directions are predominant with 38.52% and 33.24% of the total time.

Minority oversampling techniques have played a pivotal role in the field of imbalanced learning. While traditional oversampling algorithms can cause problems such as intra-class imbalance of samples, ignoring important information of boundary samples, and high similarity between new and old samples. Based on the situation, we proposed a new type of over-sampling method, BIRCH and Boundary Midpoint Centroid Synthetic Minority Over-Sampling Technique (BI-BMCSMOTE). First of all, the algorithm used the BIRCH clustering method to achieve quick cluster of the minority samples. After identifying and removing the noise, it marked the boundary minority samples in the label by probability. Secondly, it generated a density function for each sample cluster, calculated its density and sampling weight, performed midpoint composite sampling among the minority samples marked by probability and other minority samples in each cluster, and then calculated and analyzed the specific value of composite sampling to improve the accuracy of the model. According to the experimental results, the algorithm was proved to be valid.

The establishment of reliable water level prediction models is vital for urban flood control and planning. In this paper, we develop hybrid models (GA-XGBoost and DE-XGBoost) that couple two evolutionary models, a genetic algorithm (GA) and a differential evolution (DE) algorithm, with the extreme gradient boosting (XGBoost) model for hourly water level prediction. The Jungrang urban basin located on the Han River, South Korea, was selected as a case study for the proposed models. Hourly rainfall and water level data were collected between 2003 and 2020 to construct and evaluate the performance of the selected models. To compare the prediction efficiency, two other tree-based models were chosen: classification and registration tree (CART) and random forest (RF) models. A comparison of the results showed that two hybrid models, GA-XGBoost and DE-XGBoost, outperformed RF and CART in the multistep-ahead prediction of water level, and the relative errors of the hybrid model ranged from [2.18%-9.21%], compared to [3.76%-10.41%] and [2.99%-11.88%] for the RF and CART, respectively. Reliable performance was also supported by other measures. In general, the GA-XGBoost and DE-XGBoost models displayed relatively similar performance despite their small differences. The CART model was not preferable for multistep-ahead water level predictions, even though it yielded the lowest Akaike information criterion (AIC) value. This study verifies that despite having some drawbacks when considering long step-ahead prediction and model complexity, hybrid XGBoost models might be superior to many existing models for hourly water level prediction.

The existing crash modeling techniques for expressway tunnels must overcome the following difficulties: 1) The collected risk factors contributing to the tunnel crashes include narrow ranges, especially the pavement conditions and weather conditions of the tunnels are rarely taken into account. 2) Most researchers ignored the estimation deviation caused by the excess zero observations of tunnel crash datasets. 3) No existing tunnel crash model can combine the random-parameters approach and spatial-temporal approach to solve the estimation deviation caused by the inter-samples and spatial-temporal heterogeneity. To address these problems, this study presents an investigation of the safety effects of risk factors of tunnel design features, traffic conditions, pavement conditions and weather conditions utilizing a 12-quarter period (3 years) of data as well as five crash frequency models: 1) a fixed parameters negative binomial model (FPNB), 2) a random parameters negative binomial model (RPNB), 3) a random parameters negative binomial Lindley model (RPNBL), 4) a spatial and random parameters negative binomial Lindley model (SP-RPNBL), and 5) a spatial-temporal and random parameters negative binomial Lindley model (ST-RPNBL). The results showed that the ST-RPNBL model solves the deviation that arises from excess zero observations by introducing the Lindley distribution and considers the unobserved heterogeneity by introducing both the random parameters and spatial-temporal parameters that provided better goodness of fit and offered more insights into the factors that contribute to tunnel safety. Furthermore, the ST-RPNBL model detected 16 variables that were significantly correlated with tunnel crash frequency, of which 12 variables were associated with a higher crash frequency and four variables were associated with a lower crash frequency. The random variables of the curvature, the steep downgrade indicator, the proportion of class 5 vehicle and the skidding resistance index (SRI) were identified, and the influence of each significant variable on the crash frequency was analyzed.

To identify suitable merging methods to improve regional precipitation estimates using multiple sources of precipitation data, this study applied four different approaches (multiple linear regression (MLR), feedforward neural network (FNN), random forest (RF) and long short-term memory network (LSTM)) to merge four satellite precipitation products and one reanalysis data in the Jiangsu, Zhejiang and Shanghai of China. The pros and cons of the merging approaches are analyzed comprehensively, using correlation coefficient (CC), root mean square error (RMSE), relative bias (RB), probability of detection (POD), false alarm ratio (FAR), and critical success index (CSI) as evaluation indexes. Our results show that: (1) All merging approaches can improve the accuracy of precipitation estimations, but only RF and LSTM can improve the daily precipitation event detection capacity. These approaches can significantly reduce errors in moderate precipitation scenarios, but do not effectively improve accuracy in light and heavy precipitation scenarios. (2) MLR was the least expensive computing cost method in our study and performed better than the other three methods when gauge density was low. However, MLR had the worst daily precipitation event detection capacity (CSI = 0.67). (3) FNN performed moderately in most experiments (CC = 0.87, RMSE = 4.65 mm/day, RB = 1.19 %, POD = 0.94, FAR = 0.29, CSI = 0.70). (4) The merged data generated by RF was the most accurate and had the best daily precipitation event detection capacity (CC = 0.87, RMSE = 4.61 mm/day, RB = − 0.33 %, POD = 0.97, FAR = 0.20, CSI = 0.78). RF performed best in moderate precipitation scenarios. However, it performed worse than other methods when gauge density was low. (5) LSTM was the most robust methods and performed best in light precipitation scenarios. The FAR of the LSTM-generated data was the smallest (0.15) among four fusion methods. However, LSTM had the most expensive computing cost and the worst accuracy of the merged data (CC = 0.86, RMSE = 4.68 mm/day, RB = − 9.36 %).

Speech quality is often measured via subjective testing, or with objective estimators of mean opinion score (MOS) such as ViSQOL or POLQA. Typical MOS-estimation frameworks use signal level features but do not use language features that have been shown to have an effect on opinion scores. If there is a conditional dependence between score and language given these signal features, introducing language and rater predictors should provide a marginal improvement in predictions. The proposed method uses Bayesian models that predict the individual opinion score instead of MOS. Several models that test various combinations of predictors were used, including predictors that capture signal features, such as frequency band similarity, as well as features that are related to the listener, such as a language and rater index. The models are fit to the ITU-T P. Supplement 23 dataset, and posterior samples are drawn from distributions of both the model parameters and the resulting opinion score outcomes. These models are compared to MOS models by integrating over posterior samples per utterance. An experiment was conducted by ablating different predictors for several types of Bayesian hierarchical models (including ordered logistic and truncated normal individual score distributions, as well as MOS distributions) to find the marginal improvement of language and rater. The models that included language and/or rater obtained significantly lower errors (0.601 versus 0.684 root-mean-square error (RMSE)) and higher correlation. Additionally, individual rater models matched or exceeded the performance of MOS models.

Finding the nearest parking location in road networks is one of the most commonly faced challenges in everyday life of green transportation. A main challenge faced by the state-of-the-art existing parking allocation methods is to optimally offer the nearest parking location for a group of $m$ users at the cost of minimal overall traveling time to ensure the traffic and environmental sustainability. In this article, we model it as a Multiple Nearest Parking Location Allocation (MNPLA) problem, and devise a spatial index tree, called SCP-tree, to accelerate the nearest parking location allocation within the users’ time constraints. During the search process in SCP-tree, we build a pruning strategy relevant to the Geographical Preference Estimation, travel time and parking capacity to determine which branch to visit so that the search accuracy can be improved. Considering the users’ behaviors are often impacted by the geographical location and some personalized attribute information, we set the user priority based on them to help the parking officer determine the allocation sequence. We evaluate our allocation scheme using large real-world dataset with on-street parking sensor data, and extensive experimental results reveal (i) a minimum improvement of 15.9%, 1.4%, 96.9%, 160% in parking allocation time, average traveling time, I/O cost and service utility compared to the progressive methods, and (ii) a minimum improvement of 8.9%, 11.1%, 78.2%, 714% in parking allocation time, average traveling time, I/O cost and service utility compared to the baseline methods.

Diabetes mellitus and its complication such as heart disease, stroke, kidney failure, etc. is a serious concern all over the world. Hence, monitoring some important blood parameters non-invasively is of utmost importance, that too with high accuracy. This paper presents an in-house developed system, which will be helpful for diabetes patients with Chronic Kidney Disease (CKD) to monitor blood urea and glucose. This manuscript discusses a comparative study for the prediction of blood urea and glucose using Backpropagation Artificial Neural Network (BP- ANN) and Partial Least Square Regression (PLSR) model. The NVIDIA Jetson Nano board controls the five fixed LED wavelengths in the Near Infrared (NIR) region from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$2.0~\mu \text{m}$\end{document} to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$2.5~\mu \text{m}$ \end{document} with a constant emission power of 1.2 mW. The spectra for 57 laboratory prepared samples conforming with major blood constituents of the blood sample were recorded. From these samples, 53 spectra were used for training/calibration of the BP-ANN/PLSR model and the remaining 4 samples were used for validating the model. The PLSR model predicts blood urea and glucose with a Root Mean Square Error (RMSE) of 0.88 & 12.01 mg/dL, Coefficient of Determination R2 = 0.93 & R2 = 0.97, Accuracy of 94.2 % and 90.14 %, respectively. To improve the prediction accuracy, BP-ANN model is applied. Later the Principal Component Analysis (PCA) technique was applied to these 57 spectra values. These PCA values were used to train and validate the BP-ANN model. After applying the BP-ANN model, the prediction of blood urea & glucose improved remarkably, which achieved RMSE of 0.69 mg/dL, R2 = 0.96, Accuracy of 95.96 % for urea and RMSE of 2.06 mg/dL, R2 = 0.99, and Accuracy of 98.65 % for glucose. The system performance is then evaluated with Bland Altman analysis and Clarke Error Grid Analysis (CEGA).

An accurate prediction of water quality (WQ) related parameters is considered as pivotal decisive tool in sustainable water resources management. In this study, five different ensemble machine learning (ML) models including Quantile regression forest (QRF), Random Forest (RF), radial support vector machine (SVM), Stochastic Gradient Boosting (GBM) and Gradient Boosting Machines (GBM_H2O) were developed to predict the monthly biochemical oxygen demand (BOD) values of the Euphrates River, Iraq. For this aim, monthly average data of water temperature (T), Turbidity, pH, Electrical Conductivity (EC), Alkalinity (Alk), Calcium (Ca), chemical oxygen demand (COD), Sulfate (SO4), total dissolved solids (TDS), total suspended solids (TSS), and BOD measured for ten years period were used in this study. The performances of these standalone models were compared with integrative models developed by coupling the applied ML models with two different feature extraction algorithms i.e., Genetic Algorithm (GA) and Principal Components Analysis (PCA). The reliability of the applied models was evaluated based on the statistical performance criteria of determination coefficient (R2), root mean square error (RMSE), mean absolute error (MAE), Nash-Sutcliffe model efficiency coefficient (NSE), Willmott index (d), and percent bias (PBIAS). Results showed that among the developed models, QRF model attained the superior performance. The performance of the evaluated models presented in this study proved that the developed integrative PCA-QRF model presented much better performance compared with the standalone ones and with those integrated with GA. The statistical criteria of R2, RMSE, MAE, NSE, d, and PBIAS of PCA-QRF were 0.94, 0.12, 0.05, 0.93, 0.98, and 0.3, respectively.

Maize nitrogen uptake map can give growers a good opportunity in order to have valuable information regarding the efficiency of nitrogen usage in their field. The spectral information of Sentinel-2 satellite data can be used for estimating maize nitrogen uptake. In this article, Sentinel-2 data as an efficient tool were used to assess the usage of vegetation indices over three years (2017–2019), in three different locations and growing conditions to compute crop nitrogen uptake. These types of data can be used for developing an array of precision agriculture applications. Three different farms located in various climate conditions have opted for this research. For each farm, ten reference points in each year were selected to estimate a maize nitrogen uptake predictive model and use for evaluation procedure (30% of field data was for accuracy assessment and the rest of them for model prediction). At peak greenness (peak biomass) date, eight spectral vegetation indices were used for determining maize nitrogen uptake. Among these vegetation indices, simple ratio red-edge had the highest performance. It could be confirmed from the highest R 2 and the lowest root-mean-squared error (RMSE) values ( ${R^2} = \text{0.91}$ and ${\rm{RMSE}} = \text{11.34 kg/ha}$ ). Implementing this model in three different sites under various conditions proved the most top performance and accuracy of it. As a result, using near-infrared and red-edge bands in vegetation indices would be the better predictor for maize nitrogen uptake.

There exists a time lag between short-term exposure to fine particulate matter (PM2.5) and incidence of respiratory diseases. The quantification of length of the time lag is significant for preparation and allocation of relevant medical resources. Several classic lag analysis methods have been applied to determine this length. However, different models often lead to distinct results and which one is better is subtle. The prerequisite of obtaining the reliable length is that the model can truly reveal the underlying pattern hidden in the above relationship. Long short-term memory (LSTM) is an artificial recurrent neural network (RNN) architecture used in the field of deep learning, whose strong capability makes it widely applied in many fields. In this study, we manage to exploit it to acquire the time-lag length in the exposure-response relationship. The relationship between exposure and response is assumed as linear and non-linear, and models with and without confounding factors are performed under these two assumptions. Results of DLNM model show that the best hospital emergency visit prediction appears in 3 lag days, with the maximum RR value of 1.004357 (95% CI: 1.000938-1.009563). Then, a vary of LSTM models with different time steps are performed, which are evaluated by mean absolute error (MAE), the mean absolute percentage error (MAPE), the root of mean square error (RMSE) and R square (R2). The results show that LSTM of time step 3 achieves the lowest MAE (33), MAPE (9.86), RMSE (42) and the highest R2 (0.78), consistent with the result of DLNM model. Also, the proposed model is compared with ARIMA model, one of the commonly used forecasting models, showing better accuracy. This demonstrates that LSTM can be used as a new method to detect the lag effect of PM2.5 on respiratory diseases.

Snow avalanche as a natural disaster severely affects socio-economic and geomorphic processes through damaging ecosystems, vegetation, landscape, infrastructures, transportation networks, and human life. Modeling the snow avalanche has been seen as an essential approach for understanding the mountainous landscape dynamics to assess hazard susceptibility leading to effective mitigation and resilience. Therefore, the main aim of this study is to introduce and implement an ensemble machine learning model of random subspace (RS) based on a classifier, functional tree (FT), named RSFT model for snow avalanche susceptibility mapping at Karaj Watershed, Iran. According to the best knowledge of literature, the proposed model, RSFT, has not earlier been introduced and implemented for snow avalanche modeling and mapping over the world. Four benchmark models, including logistic regression (LR), logistic model tree (LMT), alternating decision tree (ADT), and functional trees (FT) models were used to check the goodness-of-fit and prediction accuracy of the proposed model. To achieve this objective, the most important factors among many climatic, topographic, lithologic, and hydrologic factors, which affect the snow accumulation and snow avalanche occurrence, were determined by the information gain ratio (IGR) technique. The goodness-of-fit and prediction accuracy of the models were evaluated by some statistical-based indexes including, sensitivity, specificity, accuracy, kappa, and area under the ROC curve, Friedman and Wilcoxon sign rank tests. Results indicated that the ensemble proposed model (RSFT), had the highest performance (Sensitivity = 94.1%, Specificity = 92.4%, Accuracy = 93.3%, and Kappa = 0.782) rather than the other soft-computing benchmark models. The snow avalanche susceptibility maps indicated that the high and very high susceptibility avalanche areas are located in the north and northeast parts of the study area, which have a higher elevation with more precipitation and lower temperatures.

This paper proposes a generalized grey target decision method (GGTDM) with index and weight both containing mixed attribute values based on Kullback-Leibler (K-L) distance. The proposed approach builds the weight function converting the mixed attribute-based weights into the certain number-based weights and takes the comprehensive weighted K-L distance as the decision-making basis (DMB). The proposed approach conducts its task in the following steps. First, all indices of alternatives are converted into binary connection numbers. Second, the two-tuple (determinacy, uncertainty) numbers originated from index binary connection numbers are obtained. Third, the two-tuple (determinacy, uncertainty) numbers of target center are calculated. Following that the certain number-based weights are obtained by the weight function. Then the comprehensive weighted K-L distance of each alternative and its target center is calculated. And the final decision making is based on the value of comprehensive weighted K-L distance with which the smaller the better. A case study illustrates the proposed approach with its effectiveness of converting the uncertain weights into the certain weights and the accurate results comparing with other decision-making methods.

The feasibility of aerial fertilization by UAV-based granular fertilizer spreaders (GFSs) has been widely accepted. Although commercial UAV-based spreaders have emerged, their performance in granular fertilizer spreading application has rarely been studied. This paper tested the performances of four existing UAV-based GFSs in the automatic controllability, range of the discharge rate and the swath width and uniformity of the distribution. An analytic hierarchy process (AHP)-based method was proposed to evaluate their comprehensive performance in spreading application. The main results are as follows: (a) GFS-A showed the maximum controllable discharge rate of 27 kg/min, and GFS-B showed the minimum (proximately 3 kg/min). The discharge rate could be changed in a stepless way by GFS-B, C and D. The relationship between the discharge rate and the rotational speed of the fluted roller of GFS-D showed a significant linear relationship ( $\text {R}^{2} = 0.9991$ ) for the operational range. (b) The distribution patterns of GFS-A and D were trapezoidal, which was the better deposition pattern for generating uniform spreading; those of the other two UAV-based GFSs were a wide triangle with obviously asymmetrical sides (GFS-B) and a typical M-shape with two peaks (GFS-C). (c) the simulation of CV-swath width illustrated that the available effective swath width (CV ≤ 20%) for these UAV-GFSs were 5.1m (CV = 14.73%), 4.8m(CV = 17.55%), 3.9m(CV = 13.02%) and 5.7m(CV = 18.36%) respectively under the given conditions. Furthermore, according to the AHP-based evaluation model, it was proved that GFS-D was able to provide superior performance in precise fertilization for a large application range, which is important for its practical application. This comparative study makes useful contribution for farmers and researchers in this field in understanding the application performance of various UAV-based GFSs.

Extreme value modelling of peak load process is critical to the reliable specification of power generation, distribution and maintenance purposes during both peak and off-peak periods. In this study, a frequency assessment of extreme peak electricity demand for the four seasons of the year using South African data for the period, January 1997 to December 2013 is carried out. A point process approach from extreme value theory is proposed as an ingenious extreme value theory approach. The data are made stationary by using a time-varying threshold which has a positive shift factor. The non-linear detrended datasets are then grouped into summer, spring, winter and autumn according to the calendar dates in the Southern Hemisphere. The datasets were declustered to keep the series relatively independent using Ferro and Segers automatic declustering method. A stationary point process model is then fitted to each of the cluster maxima. The modelling framework, which is easily extensible to other peak load parameters, assumes that peak power follows a Poisson point process. The parameters of the developed model are estimated using the maximum likelihood method. Empirical results show that daily peak electricity demand could be experienced approximately 27, 16, 7 and 15 days per year in winter, spring, summer and autumn, respectively. The modelling approach could assist system operators of utility companies in scheduling maintenance of generating units including long term planning.

The reliable prediction of dissolved oxygen concentration (DO) is significantly crucial for protecting the health of the aquatic ecosystem. The current research employed four different single AI-based models, namely long short-term memory neural network (LSTM), extreme learning machine (ELM), Hammerstein-Weiner (HW) and general regression neural network (GRNN) for modeling the DO concentration of Kinta River, Malaysia using available water quality (WQ) parameters. Afterwards, the first scenario used four different ensemble techniques (ET). Two linear, i.e. simple averaging ensemble (SAE) and weighted averaging ensemble (WAE) and two nonlinear namely; backpropagation neural network ensemble (BPNN-E) and HW ensemble (HW-E). The second scenario employed a hybrid random forest (RF) ensemble in order to enhance the prediction accuracy of the single models. The WQ parameters were subjected to a different pre-analysis test to ascertain their stability. The four-model combinations are generated using the nonlinear sensitivity input selection approach. The modeling performance was assessed using the statistical measures of Nash-Sutcliffe coefficient efficiency (NSE), Willmott's index of agreement (WI), root mean square error (RMSE), mean absolute error (MAE) and mean square error (MSE) and correlation coefficient (CC). The results of the single AI-based models demonstrated that HW (M3) served as the best model for predicting DO concentration. For ensemble results, BPNN-E (WI = 0.9764) was superior to the other three ET with average decreased of more than 2% with regards to MAE. Investigation on the hybrid RF ensemble demonstrated the reliable accuracy for all the hybrid models with better predictive skill shown by the HW-RF (CC = 0.981) ensemble. The overall results verified the promising impact of HW-M3, ET and hybrid RF ensemble for the prediction of the DO concentration in the Kinta River, Malaysia.

During preliminary tea processing, moisture content is an important consideration affecting the tea quality. Traditionally, the moisture content of tea leaves was manually controlled by the joint action of multiple processing units, and maintaining stability was difficult. In this paper, a multi-unit collaborative strategy was proposed for controlling moisture content in preliminary tea processing. Multivariate methods including polynomial regression, radical basis function neural network (RBFNN), and least squares support vector machine (LSSVM) were used to establish models for moisture content prediction in the first fixation, second fixation, and drying units, with minimal root mean square errors (RMSEs) of 1.34%, 0.86%, and 0.13%, respectively. The combination of RBFNN and LSSVM, with a RMSE of 0.03%, was used to model the preliminary processing of whole tea. Rough set data mining technology was used to obtain the optimum ranges of moisture content and critical process parameters. Finally, a Monte Carlo simulation experiment was carried out within the optimum range, and moisture content design spaces for the single unit and the whole processing line were obtained. With the proposed approach, the stability of the final moisture content of tea can be improved, which is of great significance for improving tea quality and accelerating the automation of tea production.

Control charts are a type of statistical tool used to control a production process in order to obtain the quality products that can fulfill the demands of both the manufacturer and the consumers. In this paper, we propose the Tukey Moving Average-Exponentially Weighted Moving Average control chart (MME-TCC) to detect the change of average of the process with symmetric and asymmetric distribution and to compare the efficiency in detecting the change of the MME-TCC to the MA, MME, MEM, MA-TCC and MEM-TCC at the various change levels of the parameter. The criteria to measure the efficiency were average run length (ARL), standard deviation of run length (SDRL), and median run length (MRL) which evaluated by using Monte Carlo simulation (MC), The research results showed that the proposed control chart has the highest efficiency in detecting the change when the change level was at -0.75 ≤ δ ≤ 0.75. However, if the change of parameter increased (δ ≥ 1.00), the MME had more efficiency. In the case where the observation was logistic distributions, the MA-TCC had more efficiency to detect the change. Moreover, from applying the proposed control chart to two sets of real data, the mine explosion period in the UK during 1875-1951 and data of diameter of the workpiece from an industrial factory, it was found that the MME-TCC was able to more quickly detect the change than the other control charts.

This study focuses on modeling, prediction, and analysis of confirmed, recovered, and death cases of COVID-19 by using Fractional Calculus in comparison with other models for eight countries including China, France, Italy, Spain, Turkey, the UK, and the US. First, the dataset is modeled using our previously proposed approach Deep Assessment Methodology, next, one step prediction of the future is made using two methods: Deep Assessment Methodology and Long Short-Term Memory. Later, a Gaussian prediction model is proposed to predict the short-term (30 Days) future of the pandemic, and prediction performance is evaluated. The proposed Gaussian model is compared to a time-dependent susceptible-infected-recovered (SIR) model. Lastly, an analysis of understanding the effect of history is made on memory vectors using wavelet-based denoising and correlation coefficients. Results prove that Deep Assessment Methodology successfully models the dataset with 0.6671%, 0.6957%, and 0.5756% average errors for confirmed, recovered, and death cases, respectively. We found that using the proposed Gaussian approach underestimates the trend of the pandemic and the fastest increase is observed in the US while the slowest is observed in China and Spain. Analysis of the past showed that, for all countries except Turkey, the current time instant is mainly dependent on the past two weeks where countries like Germany, Italy, and the UK have a shorter average incubation period when compared to the US and France.

Highway tunnels have a higher risk of crashing than open roads, which require a systematic approach to tunnel safety. However, previous research had the following problems: 1) Studies have largely focused on open roads, with very little research on tunnels. 2) The collected crash contributing factors involve narrow ranges, with very little tunnel crash data including both tunnel design features, traffic conditions and pavement conditions. 3) None of the studies considered both excess zero observations and unobserved heterogeneity with its interactions. To address these issues, this paper first established an appropriate tunnel dataset containing 3 to 5 years of crash data from several highways in China and the influence factors of tunnel design features, traffic conditions and pavement conditions. A correlated random parameters negative binomial Lindley (CRPNB-L) model that considers both excess zero observations and unobserved heterogeneity with its interaction effects was then proposed. Compared to the uncorrelated random parameters negative binomial Lindley (URPNB-L) model, fixed parameters negative binomial Lindley (FPNB-L) model and fixed parameters negative binomial (FPNB) model, the CRPNB-L model solves the deviation that arises from excess zero observations by introducing the Lindley distribution and considers the unobserved heterogeneity with its interactions by introducing correlated random parameters. In the comparisons, the CRPNB-L model achieves the best effects in the goodness-of-fit. Furthermore, the estimated results of the CRPNB-L model showed that segment length, traffic volume, proportion of class 5 vehicle (heavy trucks and trailers), tunnel entrance and exit segments, and steep uphill and downhill segments were associated with higher crash frequency, while curvature, tunnel length, pavement damage condition index (PCI) and skid resistance index (SRI) were associated with lower crash frequency. In addition, the random variables of the curvature, the steep downgrade indicator, the proportion of class 5 vehicle and SRI were identified and their intercorrelations were analyzed.

Postprandial blood glucose prediction is a crucial part of diabetes management. Recently, this topic has been of great interest, resulting in many research projects and published papers. Although different input parameters that might be beneficial for blood glucose prediction models were comprehensively discussed, specific data preprocessing, feature engineering and model tuning steps were not explained in detail in many of these papers. In this work, we developed and comprehensively described a data-driven blood glucose model based on a decision tree gradient boosting algorithm to predict different characteristics of postprandial glycemic responses; the model utilized meal-related data derived from a mobile app diary (including information on the glycemic index), food context (information on previous meals), characteristics of the individual patients and patient behavioral questionnaires. A set of rules was defined and implemented to detect incorrect meal records and to filter faulty data, and analyses were conducted on the overall food diary data and in particular, the data on the current meal for which the postprandial blood glucose response was calculated. Different gradient boosting models were trained and evaluated with parameters selected via random search cross-validation. The best models for the prediction of the incremental area under the blood glucose curve two hours after food intake had the following characteristics: R = 0.631, MAE = 0.373 mmol/L*h for the model not using data on current blood glucose; R = 0.644, MAE = 0.371 mmol/L*h for the model using data on the current blood glucose levels; and R = 0.704, MAE = 0.341 mmol/L*h for the model utilizing data on the continuous blood glucose trends before the meal. The impact of features was evaluated using Shapley values. The meal glycemic load, amount of carbohydrates in the meal, type of meal (e.g., breakfast), amount of starch and amount of food consumed 6 hours before the current meal were the most important contributors in the models.

Residents clustering in different periods of load fluctuation and aggregated forecasting can increase the load prediction accuracy. But the strength of load fluctuation reflects the difference in the electricity consumption behavior of residents and affects the cluster results of residents. This paper presents a new day-ahead aggregated load-forecasting method for distribution networks based on the load fluctuation and feature importance (FI) profile clustering of residents. First, the input features are determined, the FI profile of residents is determined, and residents are clustered according to the FI profile. Then, the crow search algorithm is used to optimize the initial cluster centers for preventing the clustering results from falling into a local optimum. And the cluster verification index S_Dbw, the sum of the average scattering for the clusters and the inter-cluster density, is used to evaluate the cluster quality. The optimal clustering results of the aggregated load for different fluctuation periods are determined via statistical experiments. Finally, a random forest predictor based on ensemble learning is selected. According to the optimal clustering results in different fluctuation periods, a rolling forecasting model is constructed to realize day-ahead aggregated load forecasting in a residential distribution network.

Large-scale data presents great challenges to data analysis due to the limited computer storage capacity and the heterogeneous data structure. In this article, we propose a distributed expectile regression model to resolve the challenges of large-scale data by designing a surrogate loss function and using the Iterative Local Alternating Direction Method of the Multipliers (IL-ADMM) algorithm, which is developed for the calculation of the proposed estimator. To obtain nice performance only after fewer rounds of communications, the proposed method only needs to solve an M-estimation problem on the master machine while the other working machines only to compute the gradients based on local data. Moreover, we show the consistency and the asymptotic normality of the proposed estimator, and illustrate the efficient proof by numerical simulations and positive analysis on the superconductor data.

Time Between Events (TBE) charts have advantages over the traditional control charts when monitoring high quality processes with very low defect rates. This article introduces a new discrete TBE control chart following discrete Weibull distribution. The design of the proposed chart is derived analytically and discussed numerically. Moreover, the performance is assessed by using the Average Run Length (ARL) and the Coefficient of Variation of Run Length (CVRL). Besides simulation studies, the proposed scheme is also illustrated using four real data examples.

Rainfall is a natural demolishing phenomenon. On the other side, it also serves as a major source of water when conserved through proper channel. For this issue, estimation of rain fall is of at utmost importance. The present study employed on rain fall forecasting in annual as well as non-moon session in Odisha (India). The total annual rainfall and relative humidity data were collected from period 1991-2015 from Department of Forest and Environment Govt. of Odisha. Support Vector Regression and Multilayer perception implemented for prediction of maximum rainfall in annual and non-monsoon session. Input parameter like average temperature in month, wind velocity, humidity, and cloud cover was conceder for predicting rainfall in non-monsoon session. The performance of the results was measure with MSE (mean squared error), correlation coefficient, coefficient of efficiency and MAE (mean absolute error). The results of SVR were compared to those of MLP and simple regression technique. MLP being a computationally intensive method, SVR could be used as an efficient alternative for runoff and sediment yield prediction under comparable accuracy in predictions.SVR-MLP may be used as promising alternative forecasting tool for higher accuracy in forecasting and better generalization ability.

In this work, we develop a new method for estimating the mean and standard deviation of normally distributed populations when errors exist in the sample. The proposed estimators are asymptotically unbiased if all errors are outlying errors (i.e. all errors lie outside a region defined by the parameters of normal distribution). However, the proposed method requires having an upper bound for the percentage of errors in the sample. The proposed method is compared with the already existing and commonly used estimators in terms of bias, mean square error, and Pitman closeness criterion. The proposed method is found to be superior to the others when the sample size is not too small. A simulation study is designed to test the performance of proposed estimators when most but not all errors are outlying errors. The findings of the simulation study also indicate the superiority of the proposed method when the sample size is moderately large. As an application, we use our method in Phase I of designing a control chart to improve its performance. We apply the method on a dataset where the robustness of our proposed method is tested (and compared with the other estimators) against the presence of outlying errors in Phase I data. In the findings of the application, we notice that proposed estimators were the only ones that identified out of control data points in Phase II when Phase I samples are contaminated with the errors.

To study the high quality processes, time-between-events (TBE) control charts have several advantages over the ordinary control charts. However, the existing TBE charts are based on the exponential distribution, which limit the application of these charts to monitor rare events. Therefore, to generalize existing exponential TBE charts, recently two new TBE charts assuming Weibull and generalized exponential distributions for the inter-arrival times of the renewal process, have been proposed in the literature. The main aim of this study is to present a detailed comparison of these charts. We use the expected quadratic loss (EQL), relative average run length (RARL), continuous rank probability score (CRPS) and closeness measure to assess the performance of the Weibull and the generalized exponential charts. We also discuss four real data examples in this article.

High dimensionality problem in spectra datasets is a significant challenge to researchers and requires the design of effective methods that can extract the optimal variable subset that can improve the accuracy of predictions or classifications. In this study, a hybrid variable selection method, based on the incremental number of variables using bootstrapping soft shrinkage method (BOSS) and interval random variable selection (IRVS) method is proposed and named BOSS-IRVS. The BOSS method is used to determine the informative intervals, while the IRVS method is used to search for informative variables in the informative interval determined by BOSS method. The proposed BOSS-IRVS method was tested using seven different public accessible near-infrared (NIR) spectroscopic datasets of corn, diesel fuel, soy, wheat protein, and hemoglobin types. The performance of the proposed method was compared with that of two outstanding variable selection methods i.e. BOSS and hybrid variable selection strategy based on continuous shrinkage of variable space (VCPA-IRIV). The experimental results showed clearly that the proposed method BOSS-IRVS outperforms VCPA-IRIV and BOSS methods in all tested datasets and improved the percentage of the prediction accuracy, by 15.4 and 15.3 for corn moisture,13.4 and 49.8 for corn oil, 41.5 and 50.6 for corn protein, 12.6 and 5.6 for soy moisture, 0.6 and 6.3 for total diesel fuel, 19.9 and 14.3 for wheat protein, and 5.8 and 20.3 for hemoglobin. This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/

With lifestyle and environmental changes, the prevalence of cardiovascular diseases (CVDs) is trending upwards, putting pressure on the limited medical resources. Accurate forecasting of daily counts of hospital admissions (HAs) for CVDs is helpful to optimize medical resources. In this study, we proposed a stacking ensemble model with direct prediction strategy to predict the daily number of CVDs admissions using HAs data, air pollution data, and meteorological data. The sequential forward floating selection method with early stopping was applied for feature selection. Five machine learning models, including linear regression (LR), support vector regression (SVR), extreme gradient boosting (XGBoost), random forest (RF), and gradient boosting decision tree (GBDT), were utilized as base learners to construct the stacking model. We compared the performance of the proposed stacking model with the five base learners in three datasets. The experimental results indicated that our model performed best in three datasets under four evaluation criteria, including mean absolute error (MAE), root mean square error (RMSE), mean absolute percentage error (MAPE), and coefficient of determination (R2). Particularly, in the CVDs dataset, the MAPE is 15.103 for LR, 11.862 for SVR, 10.571 for XGBoost, 10.378 for GBDT, 10.333 for RF, and 9.679 for the stacking model. Compared with the best base learner RF, the MAPE, RMSE, and MAE of the stacking model decreased by 6.3%, 7.4%, and 6.3%, respectively, and the R2 improved by 1.7%. It is evident that the proposed stacking model can effectively forecast the daily number of hospitalizations for CVDs and provide decision support for hospital managers.

Reliability assessment of MOSA devices requires accurate estimation of life distribution parameters or reliability predictors. Estimated reliability predictors consist of reasonable indicators of failure probability and mean life trends as a result of applied stress over a given period of time. Recent literature suggests finite-value or point-based estimation approach of reliability predictors in the context of continuously-applied distorted voltage stress to MOSA devices. However, this technique is prone to inaccuracies, and may not effectively reflect the impact of applied stress on reliability predictors. In this work, the use of contour plots is proposed as an alternative method for discrete mapping of estimated reliability predictors. Therefore, accelerated life tests - consisting of constant electro-thermal stress of: 85% of the reference voltage with and without harmonics, at the temperature of 135 °C for the time-period of 96 hours - are conducted on a set of low-voltage MOSA sourced from different manufacturers. The experimentally obtained times to failure are consistent with the two-parameter Weibull life distribution. The log-likelihood estimates on the Weibull probability density function are invoked using computational algorithms which yield the contour plots. These plots depict the reliability predictor mappings in the αβ plane. Results show that harmonic content in the applied stress causes extended peak amplitude which represents the point of highest probability of reduced reliability, and thus the highest accuracy point for the estimated shape parameter.

Quantitative structure-activity relationship (QSAR) regression models are mathematical ones which relate the structural properties of chemicals to the potencies of the biological activities of the chemicals. In QSAR models, the physical and chemical information of the molecules is encoded into quantitative numbers called descriptors. Recently, experimental test results (profiles) have been used as descriptors of chemicals. Profile QSAR 2.0 (pQSAR) model suggested by Martin et al. , is a multitask, two step machine learning prediction method with a combination of random forest regressions (RFRs) and partial least squares regression (PLSR). In pQSAR model, one fills the profile table’s missing values with RFRs and then builds PLSR using the profile predictions. Note that in the second step of the pQSAR method, PLSR’s predictor variables are profiles; so activity values, and the response variables are also activity values. Thus we can use the PLSRs to update the profile table and then repeat the second step. In this work, we propose an extended model of pQSAR generated by RFRs and PLSRs. Experiment of updating the given full initially predicted profile table by two kinds of prediction models, RFRs and PLSRs, has been conducted iteratively for the PKIS and ChEMBL data sets. Even though prediction performance of individual combination of RFRs and PLSRs varies, the average of the all possible predicted profile tables for given iteration shows better performance. This ensemble model has better prediction performance in sense of Pearson’s $R^{2}$ compared to that of the pQSAR model.

This paper focuses on the comparison of dimensionality reduction effect between LightGBM and XGBoost-FA. With respect to XGBoost, LightGBM can be built in the effect of dimensionality reduction via both Gradient-based One-Side Sampling(GOSS) and Exclusive Feature Bundling(EFB) algorithms, while XGBoost coupling with traditional dimensionality reduction tool Factor Analysis (XGBoost-FA) may also have dimensionality reduction effect. To present the empirical comparison, the prediagnosis dataset for the 2018 Kaggle competition Acute Liver Failure has been chosen as the research object. And pairwise comparison has been conducted among XGBoost, LightGBM, XGBoost-FA and LightGBM-FA. Concerning the test set, the vector (accuracy, log loss function, training time) of the above first four prediagnostic models are (0.75014, 0.569707, 10.5s), (0.75811, 0.576059,15.1s), (0.67786,0.663924,5.7s) and (0.67274,0.676019, 4.1s) respectively. It’s been found that the training time of XGBoost-FA (external dimensionality reduction) is shorter than that of LightGBM (build-in dimensionality reduction). Considering (accuracy, training time) being (0.82, 3.1s) published on Kaggle, the algorithm (logogram as K2a) is better than the four XGBoost-FA and LightGBM in both training time and accuracy. However, K2a removes more than 50% samples with missing values and only performs binary classification. For multi-class classification or data with a large number of missing values, XGBoost-FA is more suggested if higher operational time is required, while LightGBM is preferred if higher predictive accuracy is required. With XGBoost-FA or LightGBM being employed in AI medical services, doctors are more productive in diagnosis and treatment due to much more data support and less workload. Both complement each other.

Spatio-temporal gait parameters such as step width, cadence, stride length, and walking speed contribute to dynamic stability. Several studies have investigated the role of gait parameters in maintaining balance. However, in these studies, subjects were instructed to alter their gait. This intentional alteration has the potential to create error in the results, as subjects are not walking with a natural and comfortable gait. In consideration of this, the sample chosen in this study consisted of patients who had undergone a knee replacement. Such individuals naturally have gait parameters that differ from normal subjects. The primary objective of this study was to develop regression models that predict and measure gait stability in both the anterior-posterior and medio-lateral directions based on gait parameters. The maximum deviation of the extrapolated center of mass from the border of the base of support was the measure of gait stability. A forward stepwise multiple regression analysis was conducted to develop both models. In testing the goodness of fit of models, the values of coefficient of determination, standard error of estimates, and root mean square error were calculated. Both models showed sufficient values of goodness of fit. To improve walking stability and minimize falls, fall-prone people should walk with an adequate base-of-support area, and with lower cadence and speed. The results of this study contribute to an understanding of gait patterns and their relationship to walking stability and to how gait strategies might be taught in physical therapy programs to minimize the risk of falls.

Support Vector Regression (SVR) and its variants are widely used regression algorithms, and they have demonstrated high generalization ability. This research proposes a new SVR-based regressor: v-minimum absolute deviation distribution regression (v-MADR) machine. Instead of merely minimizing structural risk, as with v-SVR, v-MADR aims to achieve better generalization performance by minimizing both the absolute regression deviation mean and the absolute regression deviation variance, which takes into account the positive and negative values of the regression deviation of sample points. For optimization, we propose a dual coordinate descent (DCD) algorithm for small sample problems, and we also propose an averaged stochastic gradient descent (ASGD) algorithm for large-scale problems. Furthermore, we study the statistical property of v-MADR that leads to a bound on the expectation of error. The experimental results on both artificial and real datasets indicate that our v-MADR has significant improvement in generalization performance with less training time compared to the widely used v-SVR, LS-SVR, ε-TSVR, and linear ε-SVR. Finally, we open source the code of v-MADR at https://github.com/AsunaYY/v-MADR for wider dissemination.