Search

Rice is the staple food for Asia and a part of the Pacific, including Malaysia. However, the self-sufficiency ratio (SSR) for rice production in Malaysia has dropped from 69% in 2019 to 65% in 2021. For the continuous prosperity of Malaysia in food security, therefore this paper aims to investigate the statistically significant factors that affected the reduction of rice production based on the Cross Industry Standard Process for Data Mining (CRISP-DM) data science methodology. To pursue the principal objective of this paper, the annual rice production dataset period 1980-2019 corresponding to the atmospheric, climatic, and socio-economic factors under big data has been employed. Meanwhile, the regression-based predictive models employed in this paper are including multiple linear regression (MLR) and multiple nonlinear regression (MNLR) supervised machine learning models. The empirical analysis revealed that the MLR is superior rather than the MNLR in predicting rice production. Both supervised machine learning models consistently showed that the planted area and annual population are the statistically significant factors for the supervised machine learning models. In summary, this study is competent to support the advancing sustainability theme via the principal focus of poverty income revision. Specifically, the superior parsimonious predictive model resulting in this study is competent to beneficial for smallholder farmers in adopting plantation strategies for the future. Furthermore, this predictive model also could be beneficial to policymakers by providing early alarm insight about the national impact of recent atmospheric, climatic, and socio-economic trends on rice production. [ABSTRACT FROM AUTHOR]

To measure the circulatory system of skin using infrared light, PPG is extensively utilized. PPG has a lot of innate focal points like affordability, non-prominent in nature and goes probably as a versatile demonstrative gadget. It helps in the assessment of circulatory strain, oxygen inundation levels, heart yield and for administering distinctive other autonomic components of the body. For the amazing screening of different pathologies, PPG fills in as a huge promising framework. The improvement of blood in the vessel which spreads from the heart to the toes and fingertips is reflected by the PPG signals. In this paper, PPG signals are utilized for a singular patient who is encountering cardio vascular issues. For the PPG signals, the Firefly clusters and Dragonfly selector algorithms is utilized. In this study an in depth analysis of classifiers in detection of cardiovascular diseases (CVD) is done with the help of capnobase database. The following classifiers are used in this paper linear regression, Nonlinear Regression, Logistic regression, Bayesian linear discriminant classifier (BDLC) and Gaussian Mixture Model (GMM) and firefly. The plan and execution of a PPG are very economical and have a simple support. It is also used for the peripheral blood fluorescent and venous filling time. Results show that the Linear Regression classifier exceeds the other five classifiers in terms of accuracy (65.85%), F1 score (68.18%), MCC (0.316), Jaccard metric (51.72%), and error rate (34.15%). [ABSTRACT FROM AUTHOR]

A Gaussian process as prior in regression analysis was developing last decade. The performance of the prediction has shown promising. Recently the data have become large and dependent on each other. It also has nonlinear characteristics. This paper will discuss the extension of the Gaussian process using convolution for multivariate nonlinear regression. The simulation shows that the performance of prediction is good compared with the existed approach. [ABSTRACT FROM AUTHOR]

The purpose of this discourse is to describe the estimation methods of COBB-DOUGLAS production functional model thrash cost function and with multiplicative and additive errors. Nonlinear regression models have been a subject of an intensive investigation in select years. It is no exaggeration to say that in Econometric analysis, the COBB-DOUGLAS and constant Elasticity of substitution (CES) functions are the most frequently used non-linear type of special functions. Mohammad Zakir Hussain et al, in 2010, in their research article discussed the use of Cobb-Douglas production model on some selected manufacturing industries. Md.Moyazzem Hossain et al in 2015, in their research paper estimated the parameters of Cobb-Douglas production function with additive errors and multiplicative errors for some selected manufacturing industries and suggested the most suitable Cobb-Douglas production function to forecast the production process for some selected manufacturing industries for developing countries. Besides they investigated the efficiency of both capital and labor elasticity of the two mentioned form of Cobb-Douglas production function. In 1975, DE-Min Vu in his research paper derived the exact distribution of the indirect least squares estimators of the coefficients of the Cobb-Douglas production function within the content of a stochastic production model of Marshak-Andrews type. In 2017, RVSSN Rao et al, in their research paper, presented various estimation methods for estimating parameters of Cobb-Douglas production functional model. [ABSTRACT FROM AUTHOR]

The present paper models the mass-exchange processes of adsorption and desorption for a mixture of ground brown linseed (80%) and white mulberry (20%) which are bio-agricultural ingredients. The sorption capacity of the ready-made mixture was analyzed with reference to three temperatures – 10°C, 25°C and 40°C – and eight water activities in the 11% - 87% range. Via non-linear regression statistics, we modeled the modified three-parameter models of Chung-Pfost, Halsey, Oswin, and Henderson. Modified Halsey was found suitable for the description of the sorption characteristics of the ready-made mixture of brown linseed and white mulberry. The linear equations obtained from the linearization of the Brunauer-Emmett-Teller (ВЕT) model helped in the calculation of monolayer moisture content (MMC) with test points for water activity aw < 0.5 for adsorption in the 4.16% d.m. - 3.49% d.m. range and for desorpton in the 3.94% d.m. – 3.62% d.m. range. [ABSTRACT FROM AUTHOR]

This paper is about an incentive based demand reply in an electricity Market sector that can assemble a range of required tools. This is first built using an implementation system for dependent demand reply, which explained how load serving entities (LSE) aggregate and join required resources in order to contribute in the electricity market sector. The administrator will evaluate the data and the user's reply actions after providing the incentive to the user. If the admin found the reply to be as a variable, based on the change occurred in the analysis, He will be predicting the users reply behavior by using the algorithm called "LSTM" stands for long short term memory, The proposed prediction was verified under the simulation environment by a step up called tensor flow the simulation shows the comparison between linear or non-linear regression methods. [ABSTRACT FROM AUTHOR]

The literature on numerical methods for fitting nonlinear regression model has grown enormously in the fast five decades. An important phase in nonlinear regression problems is the exploration of the relation between the independent and dependent variables. A largely unexplored area of research in nonlinear regression models concerns the finite sample properties of nonlinear parameters. The main object of this research study is to propose some nonlinear methods of estimation of nonlinear regression models, namely Newton-Raphson method, Gauss-Newton method, Method of scoring, Quadratic Hill-Climbing and Conjugate Gradient methods. In 2005, G.E. Hovland et al (see [5]). In his research article, presented a parameter estimation of physical time-varying parameters for combined-cycle power plant models. B. Mahaboob et al. (see [6]), in their research paper, proposed some computational methods based on numerical analysis to estimate the parameters of nonlinear regression model. S.J. Juliear et al. (see [7]), in their research paper, developed the method of unscented transformation (UT) to propagate mean and covariance information through nonlinear transformations. [ABSTRACT FROM AUTHOR]

We consider the problem of Bayesian inference for parameters in non-linear regression models whereby the underlying unknown response functions are formed by a set of differential equations. Bayesian methods of inference for unknown parameters rely primarily on the posterior obtained by Bayes rule. For differential equation models, analytic and closed forms for the posterior are not available and one has to resort to approximations. We propose a two-stage Laplace expansion to approximate the marginal likelihood, and hence, the posterior, to obtain an approximate closed form solution. For large sample sizes, the method of inference borrows from non-linear regression theory for maximum likelihood estimates, and is therefore, consistent. Our approach is exact in the limit and does not need the specification of an additional penalty parameter. Examples in this paper include the exponential model and SIR (Susceptible-Infected-Recovered) disease spread model. [ABSTRACT FROM AUTHOR]

Nonlinear regression models have been a subject of an intensive investigation in recent years and they have wide uses in applied sciences namely Medicine, Forensic Science, Food Science, Information Science, Applied Ecology, Agronomy, Sports Science and Space Science. This research article mainly focuses on some important and innovative nonlinear estimation techniques of parameters of nonlinear regression models based on principles in matrix differentiation. The methods depicted here are principle of least squares, linear approximation method, and MLE estimation method. Oral Capps, Jr, (see [1]), in his research paper presented a theoretical discussion and some empirical results using maximum likely-hood (ML) method and iterative version of Zellner’s seemingly unrelated regression (IZEF) method in the estimation of a non linear system of demand equations when the disturbance terms are both contemporaneously and serially correlated. S. Neal, (see [3]), in his research article considered a discrete real-time nonlinear estimation problem using a least square criterion and derived a sequential algorithm which follows consideration of second order nonlinearities in system measurements. Besides some alternative nonlinear estimation techniques were discussed and examples were given which compare the various estimation algorithms. In 2003, Thomas Schon (see [4]), in his thesis, made a discussion on how to use convex optimization for solving the estimation problem. [ABSTRACT FROM AUTHOR]

This research article explores some misspecification tests for stochastic linear regression model viz. Durbin-Watson test, Ramsey's regression specification error test, Lagrange's multiplier test, and UTTS’ rainbow test. Any type of error occurred in the set of underlying assumptions of a stochastic linear regression model and the associated inferences lead to 'specification errors'. These errors will present particularly in specifying the error vector ε and the data matrix X. Generally specification errors are caused by the inclusion of irrelevant independent variables or exclusion of relevant independent variables in the stochastic linear regression model. Ivan Krivy et.al, in 2000 in their research article depicted two stochastic algorithms which are useful in estimating the parameters of nonlinear regression models. Russell Davidson et al. in 1984 (see [2]), in their paper, developed a simple computational procedure for performing a wide variety of model specification tests. In1993, LudgerRuschendorf et.al in their research article constructed as nonlinear regression representations of general stochastic processes so that they got a particular special regression representations on Markov chains and of certain m-dependent sequences. [ABSTRACT FROM AUTHOR]

In this paper, we concern at situations where we have to compare selected parameters from particular nonlinear models. A motivational examples of previously modelled data are given at the beginning. When comparing parameters from two models, it is possible to combine them into one model. But for three and more models, it is necessary to correct significance level at advance. Two approaches to solving this problem are stated, so-called Šidák correction and Bonferroni method. Results of these methods as well as results coming from the situation where significance level correction is omitted are compared based on simulated data. [ABSTRACT FROM AUTHOR]

Importance of analysis on datasets is very impactful in case of when raw datasets are needed to analyze for valuable insight. In this project three different datasets are chosen to analyze. As the main aim to find out which one is acceptable methodology in this analysis. For this analysis purpose various regression and soft computing methods have been applied. The main datasets which are needed to analyze are (i). Indian Population dataset. (ii). Sagar Island Tidal dataset, and (iii) West Bengal crop production dataset. All those datasets are collected from genuine sources. These datasets are needed to analysis so that the understanding of its insights can be finding out. Taking these datasets firstly preprocesses these to get the valuable information which are the informative data. The next step is analysis purpose, these all dataset are analyzed one by one to get finalized result. That means the methodologies like regression, deep learning technique have been applied to get which one is the best fit methodology on these datasets. So, the linear regression, nonlinear regression like linear, exponential, curvilinear and logarithmic equations have been used to find out corresponding errors likes root mean square error and percentage error. These errors demonstrate about the best fit methodology on the datasets. Beside regression technique also neural network has been applied. It has been found that neural network has given optimal less error than other technique when neural network has been applied. So, how all analysis have been done in consecutive way it is very important matter to observe. List of all datasets and methodologies which have been applied on these datasets are more discussed in the ‗implementation on datasets section'. Then result and conclusion section are to discuss about result and conclusion. Lastly some references are briefly detailed on the analysis topic. All the information about this analysis are seemed to be that the analysis on the datasets are a process of data analysis or data science. Using this data analysis technique it is understand that the valuable insight can be find out easily and necessarily [ABSTRACT FROM AUTHOR]

In different social orders, mixed drinks are generally utilized. A decent and sound way of life can be leaded if the utilization of liquor is negligible and periodic. Notwithstanding, with the over admission of liquor, it upsets the dynamic cycle of the individual and it prompts serious mental, physical and enthusiastic medical issues. The liquor identification level is imperative to decide the individual's ability to make a specific showing. In this study the detection of alcoholic effects in the brain function is analyzed through alcoholic EEG Signals. Fast Fourier Transform (FFT), Features are extracted from Alcoholic EEG signals which are classified through five classifiers viz., Linear Regression, Nonlinear Regression, Logistic Regression, Expectation Maximization (EM), and Bayesian Linear Discriminant classifier (BDLC). Nonlinear Regression Classifier outperforms others with accuracy of 89.85% and GDR of 79.69%. [ABSTRACT FROM AUTHOR]

In different social orders, mixed drinks are generally utilized. A decent and sound way of life can be leaded if the utilization of liquor is negligible and periodic. Notwithstanding, with the over admission of liquor, it upsets the dynamic cycle of the individual and it prompts serious mental, physical and enthusiastic medical issues. The liquor identification level is imperative to decide the individual's ability to make a specific showing. In this study the detection of alcoholic effects in the brain function is analyzed through alcoholic EEG Signals. Particle swarm optimization (PSO), Features are extracted from Alcoholic EEG signals which are classified through seven classifiers namely, Nonlinear Regression, Logistic Regression, Bayesian Linear discriminant Classifier (BDLC), Principal Component Analysis (PCA), Kernel PCA, Log Kernel PCA and Firefly classifier. The University of California, Irvine-Knowledge Discovery databases also known as the UCI KDD online database, provided the alcoholic EEG data analyzed in this study. Only ten well labeled alcoholic patients were considered in this study among 122 alcoholic and normal patients were available. The performance of the classifiers were analyzed and compared based on parameters like, Sensitivity, Specificity, Accuracy, GDR and Error rate. Results show that the Nonlinear Regression and BLDC classifier exceeds the other five classifiers in terms of accuracy of 94.011% and low error rate of 11.98%. [ABSTRACT FROM AUTHOR]

Empirical analysis is conducted to examine the nexus of energy consumption, energy intensity and carbon dioxide (CO2) emissions using the global panel dataset of 1990 to 2014. The main objectives are to compare the nexus between oil-importing versus oil-exporting countries, to examine the nexus between short versus long-run counterparts, and to reveal the asymmetric effects of oil price changes in the nexus examined. The main contribution of this study is to analyze the asymmetric effects of oil price changes using a nonlinear autoregressive distributed lags (NARDL) model. A few variables are included, which are energy consumption, energy intensity, total and sectoral CO2 emissions, and world crude oil price, in which the oil price acts as a control variable. The results show that CO2 emission affects energy consumption positively for both groups of countries, with a higher impact in oil-importing countries. Besides, a rise in CO2 emissions shrinks the energy intensity of oil-exporting countries but raises the energy intensity of oil-importing countries. The oil price has asymmetric effects on energy consumption, with larger effect from oil price increases. Oil price has a limited influence on energy intensity, while the higher oil price causes higher emissions. The oil-dependency factor matters, with oil-importing countries experiencing larger impacts from oil price changes. It is suggested that policy decision/ planning should co-exist with the environmental concern, to sustain the environmental quality and economic growth. [ABSTRACT FROM AUTHOR]

Human-like motion is often considered a key feature for intuitive human–robot interactions. In fact, this feature allows human peers to easily predict the robot's intention, which is perfectly aligned with the paradigm of collaborative industries, contributing to more human-centric and resilient industries. The one-sixth power law (1/6-PL) is well known in human motor control. In this work, the Human-like Upper-limb Motion Planner is used to generate three-dimensional (3D) movements of an anthropomorphic robotic arm. By applying direct kinematics, the position and orientation of the hand of the robot is determined. Subsequently, the respective curvature, torsion and velocity are computed. From a total of 600 movements, divided in six sessions, non-linear regression models are fitted and validated, in order to obtain the slope in the log-space of these movements. A statistical analysis of the parameters of the 1/6-PL is performed, and parametric and non-parametric tests are used to compare the results in each of the six sessions. [ABSTRACT FROM AUTHOR]

In this article, a method of variable selection (V.S) in the concept of sufficient dimension reduction (SDR), called SSIR-AL, is proposed. The SSIR-AL combines the ideas of adaptive Lasso with sliced inverse regression (SIR) to get a sparse SIR estimator. On other hand, the SSIR-AL method enables adaptive lasso to work with multi-dimensional and nonlinear regression without assuming any specific model. Through each of the simulation and analysis of real data was to prove the effectiveness of (SSIR-AL) [ABSTRACT FROM AUTHOR]

The performance of hexa methylene tetra amine (HMTA) (C6H12N4) as a corrosion inhibitor for mild steel in simulated CO2-saturated brine solution was investigated in the absence and presence of inhibitor at various HMTA concentrations, temperatures, and velocities using weight loss technique. Doehlert Experimental Design (DED) was used for designing of experiments. A second order polynomial equation with main, quadratic and interactive influences was suggested and optimized. Nonlinear regression statistical analysis was used for estimation of polynomial coefficients based on Quasi-Newton & Rosenbrock method. Inhibitor concentration showed a high positive impact in reducing the corrosion rate, while temperature and mixing velocity showed the negative effect. The optimum (minimum) corrosion rate was 18.714 gmd at 35.69 °C brine temperature, 74.47 ppm HMTA concentration, and 163.12 rpm mixing velocity. At these optimum conditions a 92% inhibition efficiency was obtained. [ABSTRACT FROM AUTHOR]

Neighborhood function, neighborhood radius and learning rate are terms used in neural networks theory to denote area of dependence between neurons and the degree of adaptation to new information correspondingly. For example, in Kohonen Self-Organizing Maps (SOM) – feed-forward neural networks which apply unsupervised learning - the neuron-winner (Best Matching Unit – BMU or Best Performing Unit - BPU) and his neighbors change their weights in order to be closer to the corresponding input vector. Neighborhood here is defined through neighbor function and neighbor radius. The learning rate is used when changing the weights of the BPU and its neighbors. It allows introducing different degree of change of the weights of the neurons in the neighborhood. In this way, different quantity of new knowledge is ensured for different neighbors. In this article, a new type of neighborhood function and learning rate function is proposed based on Switch-Time (ST) distribution. In neural networks, as well as in non-linear regression (which may be considered as a very simple case of neural network), different kinds of non-linear activation functions are used. In this article a new type of non-linear activation functions is proposed based on Switch-Time (ST) distribution and the term stopit regression is defined. ST distributions are also considered as regime capturing functions in neural networks with a smooth transition regime switching. [ABSTRACT FROM AUTHOR]

We develop criteria for the existence and uniqueness of global minima of a continuous bounded function on a noncompact set. Special attention is given to the problem of parameter estimation via minimization of the sum of squares in nonlinear regression and maximum likelihood. Definitions of local convexity and unimodality are given using the level set. A fundamental theorem of nonconvex optimization is formulated: If a function approaches the minimal limiting value at the boundary of the optimization domain from below and its Hessian matrix is positive definite at the point where the gradient vanishes then the function has a unique minimum. It is shown that the local convexity level of the sum of squares is equal to the minimal squared radius of the regression curvature. [ABSTRACT FROM AUTHOR]

We present advances in accurate, extremely rapid prediction of detonation performance for energetic molecules. These models may be integrated into larger efforts for high-throughput virtual screening, molecular optimization, or an experimentalist's selection of molecules before attempting a hazardous synthesis. Our machine learning workflow utilizes (a) a reference dataset generated from quantum mechanical calculations and the Cheetah thermochemical code, and (b) a directed message-passing neural network (D-MPNN) for nonlinear regression. The D-MPNN is a graph convolutional deep learning model best used with large datasets such as the one in this study. We create models to predict detonation velocity, detonation pressure, heat of formation, and density. Critically, prediction of the detonation properties requires absolutely no information other than the skeletal formula for a molecule. Molecules evaluated are CHNO-containing molecules from public datasets. Neural net architecture and training, including the Python workflow for parallel, automated dataset generation are discussed. The D- MPNN is also evaluated against LASSO and Kamlet-Jacobs models. [ABSTRACT FROM AUTHOR]

The non-contact acousto-thermal signature (NCATS) is a nondestructive evaluation technique with potential to detect fatigue in materials such as noisy titanium and polymer matrix composites. The determination of underlying physical mechanisms and properties may be determined by parameter estimation via nonlinear regression. The nonlinear regression analysis formulation, including the underlying models, is discussed. Several models and associated data analyses are given along with the assumptions implicit in the underlying model. The results of these analyses are discussed. [ABSTRACT FROM AUTHOR]

A transverse magnetic field was introduced to the arc plasma in the process of welding stainless steel tubes by high-speed Tungsten Inert Gas Arc Welding (TIG for short) without filler wire. The influence of external magnetic field on welding quality was investigated. 9 sets of parameters were designed by the means of orthogonal experiment. The welding joint tensile strength and form factor of weld were regarded as the main standards of welding quality. A binary quadratic nonlinear regression equation was established with the conditions of magnetic induction and flow rate of Ar gas. The residual standard deviation was calculated to adjust the accuracy of regression model. The results showed that, the regression model was correct and effective in calculating the tensile strength and aspect ratio of weld. Two 3D regression models were designed respectively, and then the impact law of magnetic induction on welding quality was researched. [ABSTRACT FROM AUTHOR]