Abstract
This study aims to develop an effective predictive model for lamb carcass C-site fat depth using the support vector machine (SVM) and to compare its performance against conventional multiple linear regression. Performance assessment involves root mean square error (RMSE), mean absolute error (MAE), https://www.w3.org/1998/Math/MathML" display="inline"> R 2 , https://www.w3.org/1999/xlink" xlink:href="https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781003400387/2749f5d6-f762-4fa3-9f51-835205c7b9b9/content/math6_1.tif"/> and adjusted https://www.w3.org/1998/Math/MathML" display="inline"> R 2 https://www.w3.org/1999/xlink" xlink:href="https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781003400387/2749f5d6-f762-4fa3-9f51-835205c7b9b9/content/math6_2.tif"/> metrics, computed through fivefold cross-validation. The analysis is conducted using Python and the sci-kit-learn library. Results highlight the linear kernel in SVM, employed with principal component analysis (PCA) transformed data, as the most successful kernel. This study concludes that SVM outperforms multiple linear regression in predictive accuracy for this context.
