A comparison of neural network models, fuzzy logic, and multiple linear regression for prediction of hatchability

۰ تومان

Application of appropriate models to approximate the performance function warrants more precise prediction and helps to make the best  decisions in the poultry industry. This study reevaluated the factors affecting hatchability in laying hens from 29 to 56 wk of age. Twenty-eight  data lines representing 4 inputs consisting of egg weight, eggshell thickness, egg sphericity, and yolk/albumin ratio and 1 output, hatchability,  were obtained from the literature and used to train an artificial neural network (ANN). The prediction ability of ANN was compared with that of  fuzzy logic to evaluate the fitness of these 2 methods. The models were compared using R2, mean absolute deviation (MAD), mean squared  error (MSE), mean absolute percentage error (MAPE), and bias. The developed model was used to assess the relative importance of each  variable on the hatchability by calculating the variable sensitivity ratio. The statistical evaluations showed that the ANN-based model predicted  hatchability more accurately than fuzzy logic. The ANNbased model had a higher determination of coefficient (R2 = 0.99) and lower residual  distribution (MAD = 0.005; MSE = 0.00004; MAPE = 0.732; bias = 0.0012) than fuzzy logic (R2 = 0.87; MAD = 0.014; MSE = 0.0004; MAPE =  2.095; bias = 0.0046). The sensitivity analysis revealed that the most important variable in the ANN-based model of hatchability was egg weight (variable sensitivity ratio, VSR = 283.11), followed by yolk/albumin ratio (VSR = 113.16), eggshell thickness (VSR = 16.23), and egg sphericity  (VSR = 3.63). The results of this research showed that the universal approximation capability of ANN made it a powerful tool to approximate  complex functions such as hatchability in the incubation process

دسته: , ,

توضیحات

ABSTRACT

Application of appropriate models to approximate the performance function warrants more precise prediction and helps to make the best  decisions in the poultry industry. This study reevaluated the factors affecting hatchability in laying hens from 29 to 56 wk of age. Twenty-eight  data lines representing 4 inputs consisting of egg weight, eggshell thickness, egg sphericity, and yolk/albumin ratio and 1 output, hatchability,  were obtained from the literature and used to train an artificial neural network (ANN). The prediction ability of ANN was compared with that of  fuzzy logic to evaluate the fitness of these 2 methods. The models were compared using R2, mean absolute deviation (MAD), mean squared  error (MSE), mean absolute percentage error (MAPE), and bias. The developed model was used to assess the relative importance of each  variable on the hatchability by calculating the variable sensitivity ratio. The statistical evaluations showed that the ANN-based model predicted  hatchability more accurately than fuzzy logic. The ANNbased model had a higher determination of coefficient (R2 = 0.99) and lower residual  distribution (MAD = 0.005; MSE = 0.00004; MAPE = 0.732; bias = 0.0012) than fuzzy logic (R2 = 0.87; MAD = 0.014; MSE = 0.0004; MAPE =  2.095; bias = 0.0046). The sensitivity analysis revealed that the most important variable in the ANN-based model of hatchability was egg weight (variable sensitivity ratio, VSR = 283.11), followed by yolk/albumin ratio (VSR = 113.16), eggshell thickness (VSR = 16.23), and egg sphericity  (VSR = 3.63). The results of this research showed that the universal approximation capability of ANN made it a powerful tool to approximate  complex functions such as hatchability in the incubation process.

INTRODUCTION

Modeling of a biological process (e.g., growth rate, egg production, and hatchability) in domestic animals and poultry is one of the most  important steps to approach maximum productivity and make economic decisions. In some cases, the effect of input variables are
not fully understood in the system, and latent relationships between inputs and output may exist. Therefore, classical methods of data  processing (e.g., multiple linear regression) may not be efficient to elicit the complicated interrelationships of biological responses and influencing  factors. Research has shown that the application of appropriate mathematical models may release more precise information on the system  output along with accurate prediction of performance (Baş and Boyacı, 2007b; Ahmadi and Golian, 2010, 2011; Savegnago et al., 2011; Mehri,  2012; Peruzzi et al., 2012).

Year: 2013

Publisher : Poultry Science Association Inc

By : M. Mehri

File Information: English Language/ 5 Page / size: 707 KB

Download: click

سال : 2013

ناشر : Poultry Science Association Inc

کاری از : M. Mehri

اطلاعات فایل : زبان انگلیسی / 5 صفحه / حجم : KB 707

لینک دانلود : روی همین لینک کلیک کنید

نقد و بررسی‌ها

هیچ دیدگاهی برای این محصول نوشته نشده است.

اولین کسی باشید که دیدگاهی می نویسد “A comparison of neural network models, fuzzy logic, and multiple linear regression for prediction of hatchability”

محصولات مرتبط

Mental Modeler A Fuzzy-Logic Cognitive Mapping Modeling Tool for Adaptive Environmental...The coordinating contracts of supply chain in a fuzzy decision environment