Cocoon Morphological Features Based Silk Quality Prediction Using XG Boost Algorithm

Silk quality is a critical determinant in the silk industry, significantly influencing the market value of silk products. This study proposes an innovative method for forecasting silk quality by leveraging cocoon morphological features through the application of the XG Boost algorithm. The morphological characteristics of the cocoon, including dimensions, shape, and colour, are identified as key indicators of silk quality. The XG Boost algorithm, known for its effectiveness in handling complex datasets, is employed for building a predictive model. A comprehensive dataset is assembled, encompassing a diverse set of cocoon morphological features obtained from a representative sample of silkworms. These features comprise cocoon size, length, width, shape parameters, and colour attributes. The dataset is then utilized to train and validate the XG Boost model, fine-tuning its parameters to ensure accurate prediction of silk quality. The outcomes of this study showcase the effectiveness of the proposed approach, demonstrating high predictive accuracy in determining silk quality based on cocoon morphological features. Moreover, the XG Boost algorithm facilitates the identification of crucial features that significantly contribute to the prediction, offering valuable insights into the factors influencing silk quality. In conclusion, the integration of cocoon morphological features and the XG Boost algorithm presents a promising avenue for accurate silk quality prediction. This research contributes to the advancement of silk production methodologies, introducing innovative technologies to the silk industry and fostering a pathway for the adoption of cutting-edge practices.