PREDICTION OF FROZEN SEMEN DOSES PRODUCTION IN DAIRY STUDS USING MACHINE LEARNING ALGORITHM

The ability to predict the frozen semen doses produced per ejaculate would be of considerable benefit for management skill, human resource, capital and time. new computing paradigm called machine learning involves in predicting dependent variable by complex non-linear relationship among independent variables. purpose this study is develop prediction model using one conventional modelling techniques Multiple Linear Regression (MLR) Artificial Neural Network (ANN) respectively. A Total 1,57,532 ejaculates data were used modelling. involved variables namely volume ejaculate, number, sperm concentration, initial motility post thaw motility. Various combinations architectural parameters employed explore optimum configuration each model. ANN (R2=90.66) was observed efficient over MLR (R2=73.52). root mean squared error (RMSE) value found lower (33.89) when compared (57.31). Hence, approach that could ejaculate.