Lowest Mean Absolute Deviation Research Articles

Doctor simulator: Delta-Age-Sex-AdaIn enhancing bone age assessment through AdaIn style transfer.

Bone age assessment assists physicians in evaluating the growth and development of children. However, deep learning methods for bone age estimation do not currently incorporate differential features obtained through comparisons with other bone atlases. To propose a more accurate method, Delta-Age-Sex-AdaIn (DASA-net), for bone age assessment, this paper combines age and sex distribution through adaptive instance normalization (AdaIN) and style transfer, simulating the process of visually comparing hand images with a standard bone atlas to determine bone age. The proposed Delta-Age-Sex-AdaIn (DASA-net) consists of four modules: BoneEncoder, Binary code distribution, Delta-Age-Sex-AdaIn, and AgeDecoder. It is compared with state-of-the-art methods on both a public Radiological Society of North America (RSNA) pediatric bone age prediction dataset (14,236 hand radiographs, ranging from 1 to 228 months) and a private bone age prediction dataset from Zigong Fourth People's Hospital (474 hand radiographs, ranging from 12 to 218 months, 268 male). Ablation experiments were designed to demonstrate the necessity of incorporating age distribution and sex distribution. The DASA-net model achieved a lower mean absolute deviation (MAD) of 3.52 months on the RSNA dataset, outperforming other methods such as BoneXpert, Deeplasia, BoNet, and other deep learning based methods. On the private dataset, the DASA-net model obtained a MAD of 3.82 months, which is also superior to other methods. The proposed DASA-net model aided the model's learning of the distinctive characteristics of hand bones of various ages and bothsexes by integrating age and sex distribution into style transfer.

Using Machine Learning Techniques to Forecast Mehram Company's Sales: A Case Study

Sales forecasting, situated at the intersection of art and science, is critical for inspiring managers toward achieving profitable outcomes. Its precision sustains production levels and capital and plays a pivotal role in the company's and its leaders' overall success and career progression. In the context of Mahram Food Industries, the challenge arises from diverse investor perspectives and the impactful nature of numerous variables. To address this, a new sales forecasting algorithm has been introduced to enhance accuracy. The aim is to predict future sales through a comprehensive approach, leveraging technical analysis, time series modeling, machine learning, neural networks, and random forest techniques. The research methodology integrates various advanced techniques to improve sales forecasting for Mahram Food Industries. Technical analysis, time series modeling, machine learning, neural networks, and random forest methods are combined to create a robust framework. The focus is on predicting sales for a future period within the artificial intelligence-based machine learning domain. The study employs metrics such as Mean Absolute Deviation (MAD), MAD Percentage (MADP), and Mean Squared Error (MSE) to evaluate and compare the performance of the proposed neural network against traditional methods like multiple variable regression and time series modeling. The study's results highlight the superior performance of the neural network in sales forecasting for Mahram Food Industries. The Mean Absolute Deviation (MAD) for the neural network is 28.33, outperforming multiple variable regression (28.54) and time series modeling (29.45). Additionally, the neural network demonstrates a better MAD Percentage (MADP) with a value of 10.2%, surpassing the values associated with multiple variable regression (10.35%) and time series modeling (10.30%). The Mean Squared Error (MSE) further confirms the neural network's superiority with a value of 6452 compared to 6472 and 7865 for multiple variable regression and time series modeling, respectively. In conclusion, the study showcases the effectiveness of integrating advanced techniques, particularly the neural network, in enhancing the accuracy of sales forecasting for Mahram Food Industries. The comprehensive approach, combining technical analysis, time series modeling, machine learning, neural networks, and random forest, is a valuable strategy for predicting future sales. The superior performance of the neural network, as evidenced by lower MAD, MADP, and MSE values, suggests its potential for guiding informed decision-making in goal setting, hiring, budgeting, and other critical aspects of business management.

Correlation and determination of the Metabolizable Energy (ME) of tropical forage with nutrient content for ruminants

The metabolizable energy (ME) of tropical forages measured by in vivo method in ruminants had a high degree of accuracy but requires a long time and is expensive. One method that can be done is the ME estimation model. The objectives of the present study were carried out to investigate the relationship between tropical forage nutrient content and ME for ruminants as well as determine and validate a model for estimating ME of tropical forage based on nutrient content. A total of 26 forage samples consisting of 14 types of grass and 12 legumes were obtained after data pre-processing or data cleaning and data normalization. Forage samples will be grouped into 3, Grass + Legume (G+L=26), grass (R=14), and legume (L=12). The database used is Crude Protein (CP), Extract Ether (EE), Neutral Detergent Fiber (NDF), Acid Detergent Fiber (ADF), and hemicellulose as well as ME with in vivo experiments. The initial stage is preprocessing data. Nutrient content and ME were analyzed using Pearson Correlation and followed by multiple linear regression to determine the ME estimation model. However, validated used the mean absolute deviation (MAD), root means square error (RMSE), and mean absolute percentage error (MAPE). The results showed that there was a significant and highly significantly correlated between nutrient composition and ME in the Grass + Legume, Grass, and Legume groups so it could be used to determine ME. There are 9 regression equations with significance and have high R2 and after being validated with the lowest MAD, RMSE, and MAPE values, three regression equations are obtained with one each for each group Grass + Legume (G+L), Grass (R), and Legumes (L). It is concluded that the regression equation of ME of tropical forage is MER+L = 12.429 – 0.122 ADF for Grass + Legume, EMR = 15.609 – 0.115 NDF for Grass, and EML = 3.726 – 0.186 CP for Legume.

Predicting Higher Heating Value of Sewage Sludges via Artificial Neural Network Based on Proximate and Ultimate Analyses

The higher heating value (HHV) was an important factor for measuring the energy recovery price of sewage sludge, which was commonly determined by oxygen bomb calorimeter; however, there were problems of time consuming and high measurement cost. In this study, a back-propagation neural network (BPNN) model based on proximate and ultimate combination analysis was developed to predict the HHV of sewage sludge and the accuracy of the model was illustrated using statistical analysis. The results showed that the BPNN model had good accuracy, with a regression coefficient of 0.979 and 0.975 for the training and test groups, respectively. Several previously proposed linear models for predicting the HHV of sewage sludge were selected for comparison. The results showed that the BPNN model was the best among all models with the highest regression coefficient (0.975) and the lowest mean absolute deviation (0.385).

Economic Representative Scenarios for Variable Annuity Dynamic Hedging of GMMB and GMDB

Variable annuities introduce significant risk for the insurers. To control this risk, insurers generally use dynamic hedging models. In practice, calculations for dynamic hedging models for variable annuities are computationally intensive since they require many nested stochastic scenario projections with outer and inner loops. In this article, we study the relationship between the hedging errors and the descriptive statistics of economic scenarios. Then, using the relationships between the hedging errors and the descriptive statistics of the economics scenarios, we present two novel scenario selection algorithms to determine economic representative scenarios (ERS) to run for dynamic risk hedging models. We do this for guaranteed minimum maturity benefit (GMMB) and guaranteed minimum death benefit (GMDB). We also benchmark these new algorithms against a control variate algorithm and Monte Carlo simulations. One of the two proposed novel scenario selection algorithms, the adapted Clara algorithm gives a lower mean absolute deviation of the conditional tail expectation (CTE) than the control variate method and Monte Carlo simulations for the same number of scenarios. Also, 500 representative scenarios selected by the adapted Clara algorithm generate a mean absolute deviation of the CTE95 of the hedging errors approximately equivalent to 1000 (or more in some cases) Monte Carlo simulations for 5 years GMMB and GMDB.

A multi-locus linear mixed model methodology for detecting small-effect QTLs for quantitative traits in MAGIC, NAM, and ROAM populations

Although multi-parent populations (MPPs) integrate the advantages of linkage and association mapping populations in the genetic dissection of complex traits and especially combine genetic analysis with crop breeding, it is difficult to detect small-effect quantitative trait loci (QTL) for complex traits in multiparent advanced generation intercross (MAGIC), nested association mapping (NAM), and random-open-parent association mapping (ROAM) populations. To address this issue, here we proposed a multi-locus linear mixed model method, namely mppQTL, to detect QTLs, especially small-effect QTLs, in these MPPs. The new method includes two steps. The first is genome-wide scanning based on a single-locus linear mixed model; the P-values are obtained from likelihood-ratio test, the peaks of negative logarithm P-value curve are selected by group-lasso, and all the selected peaks are regarded as potential QTLs. In the second step, all the potential QTLs are placed on a multi-locus linear mixed model, all the effects are estimated using expectation-maximization empirical Bayes algorithm, and all the non-zero effect vectors are further evaluated via likelihood-ratio test for significant QTLs. In Monte Carlo simulation studies, the new method has higher power in QTL detection, lower false positive rate, lower mean absolute deviation for QTL position estimate, and lower mean squared error for the estimate of QTL size (r2) than existing methods because the new method increases the power of detecting small-effect QTLs. In real dataset analysis, the new method (19) identified five more known genes than the existing three methods (14). This study provides an effective method for detecting small-effect QTLs in any MPPs.

Age prediction by methylation analysis of small amounts of DNA using locked nucleic acids.

Age prediction based on methylation analysis has been reported in many populations, with 10ng or more of DNA usually required for each determination. In this study, we designed thermostable locked nucleic acid (LNA) primers by replacing a small number of DNA bases in standard DNA primers with LNAs. We evaluated these primer sets by single-base extension analysis using 10, 5, or 2ng of DNA that would be less than template DNA used in standard methylation testing, and determined sensitivity and accuracy. We analyzed EDARADD, SST, and KLF14 genes, targeting one CpG site in each gene. Melting temperature values of most LNA primers were 4°C higher than those of DNA primers. The intensities of signals from the EDARADD and SST genes were significantly improved by the LNA primers, by 3.3 times and 1.4 times, respectively, compared with the DNA primers using 2ng of DNA. Coefficient of variation (CV) analysis was used to assess the accuracy of the determined methylation levels. CVs were increased using small amounts of DNA, but lower CVs were detected using LNA primers. We also showed high accuracy of age prediction for 51 individuals using LNA primers. The lowest mean absolute deviation was obtained using 10ng of DNA and was 3.88 years with the LNA primers. Thermostable PCR primers were simply designed, and the LNAs improved the sensitivity and accuracy of methylation analysis for 10ng or less of DNA.

Understanding the accuracy of modelled changes in freshwater provision over time

Accurate modelling of changes in freshwater supplies is critical in an era of increasing human demand, and changes in land use and climate. However, there are concerns that current landscape-scale models do not sufficiently capture catchment-level changes, whilst large-scale comparisons of empirical and simulated water yield changes are lacking. Here we modelled annual water yield in two time periods (1: 1985–1994 and 2: 2008–2017) across 81 catchments in England and validated against empirical data. Our objectives were to i) investigate whether modelling absolute or relative change in water yield is more accurate and ii) determine which predictors have the greatest impact on model accuracy. We used the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) Annual Water Yield model. In this study, absolute values refer to volumetric units of million cubic metres per year (Mm3/y), either at the catchment or hectare level.Modelled annual yields showed high accuracy as indicated by the low Mean Absolute Deviation (MAD, based on normalised data, 0 is high and 1 is low accuracy) at the catchment (1: 0.013 ± 0.019, 2: 0.012 ± 0.020) and hectare scales (1: 0.03 ± 0.030, 2: 0.030 ± 0.025). But accuracy of modelled absolute change in water yield showed a more moderate fit on both the catchment (MAD = 0.055 ± 0.065) and hectare (MAD = 0.105 ± 0.089) scales. Relative change had lower accuracy (MAD = 0.189 ± 0.135). Anthropogenic modifications to the hydrological system, including water abstraction contributed significantly to the inaccuracy of change values at the catchment and hectare scales.Quantification of changes in freshwater provision can be more accurately articulated using absolute values rather than using relative values. Absolute values can provide clearer guidance for mitigation measures related to human consumption. Accuracy of modelled change is related to different aspects of human consumption, suggesting anthropogenic impacts are critically important to consider when modelling water yield.

Optimization of ultrasound-assisted extraction of bitter compounds and polyphenols from willow gentian underground parts

• Underground parts of Gentiana asclepiadea as a source of bioactive compounds. • Optimization and modeling of the extraction process by RSM and ANN approach. • Extraction of gentiopicroside, isoorientin, isovitexin, isogentisin and polyphenols. • The experimentally obtained values were in close agreement with the predicted. • The ANN model showed superior predictive capacity compared to the RSM model. Willow gentian ( Gentiana asclepiadea L., Gentianaceae) is a medicinal plant that is primarily used in the treatment of gastrointestinal and liver diseases in folk medicine. As a bitter raw material, it is also used in the food industry for liquors flavoring. In order to optimize the ultrasound-assisted extraction (UAE) of gentiopicroside, isovitexin, isoorientin, and isogentisin, as well as total polyphenols from the underground parts of willow gentian, the influence of process parameters was investigated. Extraction time (from 5 to 65 min), ethanol concentration (from 10 to 90% v/v), solid to liquid ratio (from 1:10 to 1:50 w/v), and extraction temperature (from 20 to 80 °C) were estimated at five levels as independent variables. The response surface methodology (RSM) and the artificial neural network (ANN) approach were employed to develop a predictive extraction model and to establish optimal extraction conditions to simultaneously maximize the extraction yield of all target compounds. Established optimal values of process parameters were: extraction time 50 min, ethanol concentration 53% (v/v), solid to liquid ratio 1:40 (w/v), and extraction temperature 65 °C. The obtained extraction yields under predicted optimal conditions for gentiopicroside (49.23 mg/g DW), isoorientin (1.00 mg/g DW), isovitexin (4.04 mg/g DW), isogentisin (1.14 mg/g DW), and TPC (10.03 mg of GAE/g DW) confirmed the adequacy of both RSM and ANN models. Compared with the RSM model, the ANN model fitted better with the experimental values of each compound (higher coefficient of determination and lower mean absolute deviation) indicating its higher predictive capacity. The developed high-accuracy extraction model provides the basis for further separation and purification, as well as for studying the bioactivity of selected compounds from underground parts of G. asclepiadea .

Determination of minimum stock on system retail using forecast, economic order quantity and reorder point methods

In a globalization era, where retail business competition is getting tougher, a retail business has to have more advantages such as competitive selling prices, stock availability and others. In order to have competitive selling prices, a retail business should be able to arrange planning regarding goods availability to maintain balance between demand and existing stock (supply). The need for information on the availability of stock of goods in accordance with the sales that occur becomes very important. It is needed to avoid the goods accumulated excessively in the warehouse or even shortage of goods. If this kind of information about stocks both in a warehouse or the one that has been sold does not properly provided, it can cause reduction on the company’s profit because it can lead to more expenses on buying other goods as guessing that the stock of certain goods are running low or even more miscalculating to buy more goods that is actually still on the warehouse with sufficient quantities. These kind of situations were often occurred when a retail business do not have a tool to ensure the estimated purchase of stock needed for a certain period of time during selling. Therefore, it is required a system to help facilitate companies in carrying out retail business activities by using the Forecasting Method to help estimate quantitatively what will happen in the future with the stocks, based on relevant data in the past. In accordance with the results of the identification of demand patterns, the Forecasting Method used is Moving Average (MA), Weighted Moving Average (WMA), Single Exponential Smoothing (SES) and Double Exponential Smoothing (DES). After Forecasting, the best Forecast Method which has the lowest Mean Absolute Deviation (MAD) selection is made. This Economic Order Quantity (EOQ) method helps to determine the optimal purchase frequency. Through determining the optimal number and frequency of purchases, then optimal inventory control is obtained. The Reorder Point method can be used to find out what the minimum point is to be able to order a stock back, so that it is expected that the amount of stock that is not excessive and can meet sales needs optimally.

Congruence between Meshes and Library Files of Implant Scanbodies: An In Vitro Study Comparing Five Intraoral Scanners.

Purpose. To compare the reliability of five different intraoral scanners (IOSs) in the capture of implant scanbodies (SBs) and to verify the dimensional congruence between the meshes (MEs) of the SBs and the corresponding library file (LF). Methods. A gypsum cast of a fully edentulous maxilla with six implant analogues and SBs screwed on was scanned with five different IOSs (PRIMESCAN®, CS 3700®, MEDIT i-500®, ITERO ELEMENTS 5D®, and Emerald S®). Ten scans were taken for each IOS. The resulting MEs were imported to reverse engineering software for 3D analysis, consisting of the superimposition of the SB LF onto each SB ME. Then, a quantitative and qualitative evaluation of the deviations between MEs and LF was performed. A careful statistical analysis was performed. Results. PRIMESCAN® showed the highest congruence between SB MEs and LF, with the lowest mean absolute deviation (25.5 ± 5.0 μm), immediately followed by CS 3700® (27.0 ± 4.3 μm); the difference between them was not significant (p = 0.1235). PRIMESCAN® showed a significantly higher congruence than MEDIT i-500® (29.8 ± 4.8 μm, p < 0.0001), ITERO ELEMENTS 5D® (34.2 ± 9.3 μm, p < 0.0001), and Emerald S® (38.3 ± 7.8 μm, p < 0.0001). CS 3700® had a significantly higher congruence than MEDIT i-500® (p = 0.0004), ITERO ELEMENTS 5D® (p < 0.0001), and Emerald S® (p < 0.0001). Significant differences were also found between MEDIT i-500® and ITERO ELEMENTS 5D® (p < 0.0001), MEDIT i-500® and Emerald S® (p < 0.0001), and ITERO ELEMENTS 5D® and Emerald S® (p < 0.0001). Significant differences were found among different SBs when scanned with the same IOS. The deviations of the IOSs showed different directions and patterns. With PRIMESCAN®, ITERO ELEMENTS 5D®, and Emerald S®, the MEs were included inside the LF; with CS 3700®, the LF was included in the MEs. MEDIT i-500® showed interpolation between the MEs and LF, with no clear direction for the deviation. Conclusions. Statistically different levels of congruence were found between the SB MEs and the corresponding LF when using different IOSs. Significant differences were also found between different SBs when scanned with the same IOS. Finally, the qualitative evaluation revealed different directions and patterns for the five IOSs.

Providing accurate answers to OLAP queries based on standardized moments of data cubes

In this paper, we focus on the problem of providing accurate estimates to a target data cube from sets of source data cubes, which share the same summary measures. We investigate the acyclic and cyclic schemas of data sources and show that the more accurate target data cube can be computed on the basis of third and fourth standardized moments (i.e., skewness and kurtosis, respectively) of the source data cubes. In particular, we consider the mean absolute deviation of these parameters over the data cubes involved in the sets. For both standardized moments, we prove formally the set that achieves higher mean of skewness/kurtosis shows lower mean absolute deviation of skewness/kurtosis. On the basis of these results, we prove the set of data cubes with the lower mean absolute deviation of skewness/kurtosis achieves more accurate target estimate. We show that groups with cyclic schemas that present lower mean absolute deviation of standardized moments, achieve more accurate target estimate with respect to the ones with acyclic schemas as well. We define an algorithm that identifies within each sets of acyclic and cyclic groups, the one that achieves more accurate estimate for the target data cube on the basis of standardized moments. We provide some experimental results in order to verify our theoretical results.

Characteristics of frictional pressure drop of two-phase nitrogen flow in horizontal smooth mini channels in diabatic/adiabatic conditions

The two-phase frictional pressure drop of nitrogen under diabatic and adiabatic conditions is investigated experimentally in two horizontal mini channels. The effects of factors in a variety, including the channel diameter, mass flux, heat flux, inlet pressure and vapor quality are examined on the frictional pressure drop and liquid-phase multiplier of the two-phase flow. The two-phase frictional pressure drop predicted by previous correlations from other studies are compared with the experimental results of this study, and the lowest mean absolute deviation is 18.3% for adiabatic conditions and 10.9% for adiabatic conditions. Based on the experimental results of nitrogen flow, two new correlations in the form of the Chisholm-C correlation are developed to predict the two-phase frictional pressure dropby involving the effect of surface tension, and the mean absolute deviation of which is 12.3% and 10.4% for the adiabatic and diabatic conditions respectively. The new correlations are also shown able to well predict the experimental frictional pressure drop of other studies under different experimental conditions.

Hydrogen Activation by Silica-Supported Metal Ion Catalysts: Catalytic Properties of Metals and Performance of DFT Functionals.

Single-site heterogeneous catalysts (SSHC) have received increasing attention due to their well-defined active sites and potentially high specific activity. Detailed computational studies were carried out on a set of potential SSHC's, i.e., silica-supported metal ions, to investigate the reactivity of these catalysts with H2 as well as to evaluate the performance of density functional theory (DFT) methods in conjunction with triple-ζ quality basis sets (i.e., cc-pVTZ) on reaction energetics. The ions considered include 4d and 5d metals as well as several post-transition metal ions. A representative cluster model of silica is used to calculate reaction free energies of the metal hydride formation that results from the heterolytic cleavage of H2 on the M-O bond. The hydride formation free energy is previously shown to be strongly correlated with the catalytic activity of such catalysts for alkene hydrogenation. ONIOM calculations (CCSD(T)//MP2) are used to assess the accuracy and reliability of the MP2 results and it is found that MP2 is a suitable level of theory for gauging the performance of DFT functionals. The performance of various DFT functionals is assessed relative to MP2 results and it is found that the wB97xd and PBE0 functionals have the lowest standard deviation (STD) value while the MN12SX and PBE functionals have the lowest mean absolute deviation (MAD) values. The B3LYP functional is shown to have similar MAD and STD values as the top performing functionals. Potential active SSHC's for exergonic hydrogen activation predicted in this study include mostly late and post transition metal ions, i.e., Au3+, Pd2+, Pt4+, Pd4+, Ir4+, Hg2+, Rh3+, Pb4+, Tl3+, In3+, Ir3+, Os4+, Cd2+, Ru2+, and Ga3+. This study provides important guidance to future computational studies of such catalyst systems.

Two-stage estimation of the impact of contextual variables in stochastic frontier production function models using Data Envelopment Analysis: Second stage OLS versus bootstrap approaches

Banker and Natarajan (2008) identified sufficient conditions in a stochastic framework to justify the popular two-stage approach of Data Envelopment Analysis followed by Ordinary Least Squares (DEA + OLS) to estimate the impact of contextual variables on productivity. Simar and Wilson (2007) suggested an alternative approach involving a second-stage truncated regression model and bootstrap method to estimate confidence intervals. We show that the effectiveness of the Simar–Wilson approach critically depends on the assumption that the actual data generating process (DGP) exactly matches their assumed DGP and their approach does not yield correct inferences in environments characterized by stochastic noise. Extensive simulations from a stochastic frontier data generating process document that the simple two-stage DEA + OLS model significantly outperforms the more complex Simar–Wilson model with lower mean absolute deviation (MAD), lower median absolute deviation (MEAD) as well as higher coverage rates when the contextual variables significantly impact productivity.Subject classifications: Data Envelopment Analysis, productivity, two-stage, contextual variables, performance, nonparametric, simulationArea of review: Optimization.

Chaotic time series prediction based on brain emotional learning model and self-adaptive genetic algorithm

Chaos phenomenon is one of the most important physical phenomena, which has significant effects on one's production and life. Therefore, it is indispensable to find out the regularity of chaotic time series from a chaotic system for weather forecasting, space missions, alarm systems, etc. Although various models and learning algorithms have been developed to predict chaotic time series, many traditional methods suffer drawbacks of high computational complexity, slow convergence speed, and low prediction accuracy, due to extremely complex dynamic characteristics of chaotic systems. In this paper, a brain-inspired prediction model, i.e., brain emotional learning (BEL) model combined with self-adaptive genetic algorithm (AGA) is proposed. The establishment of BEL model is inspired by the neurobiology research, which has been put forward by mimicking the high-speed emotional learning mechanism between amygdala and orbitofrontal cortex in mammalian brain, it has advantages of lowcomputational complexity and fast learning. The BEL model employs reward-based reinforcement learning to adjust the weights of amygdala and orbitofrontal cortex. However, the reward-based method is modelsensitive and hard to generalize to other issues. To improve the performance of BEL model, AGA-BEL is proposed for chaotic prediction, in which the AGA is employed for parameter optimization. Firstly, weights and biases of orbitofrontal cortex and amygdala in BEL model are distributed to chromosomal gene sequence for optimization. Secondly, fitness function is employed to adjust the weights of amygdale and orbitofrontal cortex by self-adaptive crossover and mutation operations Therefore, the parameter optimization problem is transformed into a function optimization problem in the search space. Finally, the best chromosome that represents the best combination of weights and biases for BEL model is chosen, which is used for chaotic prediction. Prediction experiments on the benchmark Lorenz chaotic time series and a real-world chaotic time series of geomagnetic activity Dst index are performed. The experimental results and numerical analysis show that the proposed AGA-BEL prediction model achieves lower mean absolute deviation, mean square error, mean absolute percentage error, and higher correlation coefficient than the original BEL, levenberg marquardt-back propagation (LM-BP) and multilayer perceptron-back propagation (MLP-BP). Meanwhile, the BEL-based models take less computational time than the traditional BP-based models. Therefore, the proposed AGA-BEL model possesses the advantages of fast learning and low computational complexity of BEL model as well as the global optimum solution of AGA. It is superior to other traditional methods in terms of prediction precision, execution speed, and stability, and it is suited for online prediction in fast-varying environments.

Modeling Spatial Effect in Residential Burglary: A Case Study from ZG City, China

The relationship between burglary and socio-demographic factors has long been a hot topic in crime research. Spatial dependence and spatial heterogeneity are two issues to be addressed in modeling geographic data. When these two issues arise at the same time, it is difficult to model them simultaneously. A cross-comparison of three models is presented in this study to identify which spatial effect should be addressed first in crime analysis. The negative binominal model (NB), Bayesian hierarchical model (BHM) and the geographically weighted Poisson regression model (GWPR) were implemented based on a three-year residential burglary data set from ZG, China. The modeling result shows that both BHM and GWPR outperform NB as they capture either of the spatial effects. Compared to the NB model, the mean absolute deviation (MAD) of BHM and GWPR was decreased by 83.71% and 49.39%, the mean squared error (MSE) of BHM and GWPR was decreased by 97.88% and 77.15%, and the R d 2 of BHM and GWPR was improved by 26.7% and 19.1%, respectively. In comparison with BHM and GWPR, BHM fits the data better with lower MAD, MSE and higher R d 2 . The empirical analysis indicates that the percentage of renter population, percentage of people from other provinces, bus line density, and bus stop density have a significantly positive impact on the number of residential burglaries. The percentage of residents with a bachelor degree or higher, on the other hand, is negatively associated with the number of residential burglaries.

Assessment of quantum chemical methods for the calculation of homolytic N–F bond dissociation energies

Abstract In this article, the performance of a wide range of conventional and double-hybrid DFT methods (in conjunction with Dunning basis sets of double-, triple- and quadruple-zeta quality), as well as a number of Gaussian-n thermochemical protocols are assessed for their ability to compute accurate homolytic N–F bond dissociation energies (BDEs). Their performance is evaluated against a previously reported set of 31 highly accurate gas-phase N–F BDEs obtained using the benchmark-quality W2w thermochemical protocol (See: R.J. O'Reilly, A. Karton, L. Radom, J. Phys. Chem. A 2011, 115, 5496.). Out of all of the DFT/basis set combinations investigated, ωB97 and M06-2X (in conjunction with the aug'-cc-pVDZ basis set) offer the lowest mean absolute deviations (MADs = 2.4 and 2.7 kJ mol–1, respectively). Of the Gaussian-n procedures, G3X offers the best performance (MAD = 1.4 kJ mol–1), whilst the significantly more economical G3X(MP2)-RAD method also offers excellent performance (MAD = 1.8 kJ mol–1).

Macroscopic modeling of pedestrian and bicycle crashes: A cross-comparison of estimation methods

The paper presents a cross-comparison of different estimation methods to model pedestrian and bicycle crashes. The study contributes to macro level safety studies by providing further methodological and empirical evidence on the various factors that influence the frequency of pedestrian and bicycle crashes at the planning level. Random parameter negative binomial (RPNB) models are estimated to explore the effects of various planning factors associated with total, serious injury and minor injury crashes while accounting for unobserved heterogeneity. Results of the RPNB models were compared with the results of a non-spatial negative binomial (NB) model and a Poisson-Gamma-CAR model. Key findings are, (1) the RPNB model performed best with the lowest mean absolute deviation, mean squared predicted error and Akaiki information criterion measures and (2) signs of estimated parameters are consistent if these variables are significant in models with the same response variables. We found that vehicle kilometers traveled (VKT), population, percentage of commuters cycling or walking to work, and percentage of households without motor vehicles have a significant and positive correlation with the number of pedestrian and bicycle crashes. Mixed land use is also found to have a positive association with the number of pedestrian and bicycle crashes. Results have planning and policy implications aimed at encouraging the use of sustainable modes of transportation while ensuring the safety of pedestrians and cyclist.

Derivation of a Multiparameter Gamma Model for Analyzing the Residence-Time Distribution Function for Nonideal Flow Systems as an Alternative to the Advection-Dispersion Equation

A new residence-time distribution (RTD) function has been developed and applied to quantitative dye studies as an alternative to the traditional advection-dispersion equation (AdDE). The new method is based on a jointly combined four-parameter gamma probability density function (PDF). The gamma residence-time distribution (RTD) function and its first and second moments are derived from the individual two-parameter gamma distributions of randomly distributed variables, tracer travel distance, and linear velocity, which are based on their relationship with time. The gamma RTD function was used on a steady-state, nonideal system modeled as a plug-flow reactor (PFR) in the laboratory to validate the effectiveness of the model. The normalized forms of the gamma RTD and the advection-dispersion equation RTD were compared with the normalized tracer RTD. The normalized gamma RTD had a lower mean-absolute deviation (MAD) (0.16) than the normalized form of the advection-dispersion equation (0.26) when compared to the normalized tracer RTD. The gamma RTD function is tied back to the actual physical site due to its randomly distributed variables. The results validate using the gamma RTD as a suitable alternative to the advection-dispersion equation for quantitative tracer studies of non-ideal flow systems.