3789

3580

كارشناسي ارشد

علوم دامي

اصفهان : دانشگاه صنعتي اصفهان، دانشكده كشاورزي

1386

دوازده، 91، [II] ص.: مصور،جدول،نمودار

ص.ع. به فارسي و انگليسي

محمد علي ادريس

مهدي ادريس، حميد رضا رحماني

4/10/86

حسين خادمي، مسعود عليخاني

مهندسي كشاورزي

ID3580

به فارسي و انگليسي : قابل رؤيت در نسخه ديجيتال

AbstractMilk production in dairy cattle is affected by linear and non linear interactions betweengenetic and environmental effects While conventional methods are based on linearrelationships Artificial Neural Network system also considers non linear relationshipsbetween parameters In many countries analysis of milk traits for 305 days in lactationis the foundation of dairy cattle genetic evaluations Thus a mathematical model forprediction of second parity milk yield and fat percentage with the use of first parityinformation seemed to be helpful as a tool for predicting the yield of prospectiveproducing cows In this study a back propagation neural network and multiple linearregression methods were compared based on their prediction differences with observedvalues Data from 4 large sized dairy farms in Isfahan were used From 1880 availablerecords of first and second parities 1850 records were used for training testing andevaluation a back propagation artificial neural network system and multiple linearregression model and 30 randomly chosen records for simulation The obtained resultsof the simulation showed that artificial neural network with lower RMSE 817 84 0 336 for milk yield kg and fat respectively and S D ratio 0 576 0 106 for milkyield kg and fat respectively than multiple linear regressions with RMSE 933 93 0 350 for milk yield kg and fat respectively and S D ratio 0 697 0 544 for milkyield kg and fat respectively and also higher adjust coefficient of determination 36 74 72 88 for milk yield kg and fat respectively than multiple linear regression 32 70 40 35 for milk yield kg and fat respectively presented the better result Fitness of both ANN and MLR for MF relative to MY which showed that theemployed input variables may be can good explain the change in fat than milk yieldkg And also this result shows that ANNs are reliable as a decision support system thathelps breeders to choose a cow to be left or culled from herd than MLR Selection of aproper sample is very important for training an ANN The efficiency of ANNs will bemore improved when samples and variables which are more relevant to the outputvariables are used Increase in response accuracy of network with adding input andoutput variable with high correlation Thus a flexibility of this method relative to MLRis that it can be further developed for health fertility lifetime and other economicaltraits in dairy industry In MLR Adding new data requires a new statistical model whereas a neural network system can update itself with new data and ANN can beimproved with more additional input variables and trained with more actual data to getmore accurate prediction finally the results of the simulation shown that there was nosignificant difference between the observed and the predicted second parity milk yieldand fat percentage P 0 05 and The major use of this predictive process is to makeaccurate selection decisions which are based on prior knowledge of the outcomes

محمد علي ادريس

مهدي ادريس، حميد رضا رحماني

حسين خادمي، مسعود عليخاني