Sand Cat Swarm Optimizer with CatBoost for Sarcoidosis Diagnosis

In the last few years, machine learning has increased in popularity across many disciplines. This paper aims to comprehensively analyze the CatBoost classification algorithm in the context of Sarcoidosis. Analysis was undertaken to evaluate the performance of the CatBoost classification algorithm in comparison to other classifiers. The CatBoost algorithm outperformed other classifiers exploited in this study to identify and differentiate Sarcoidosis. Previous scholarly works ignored missing data observations or filled them with mean values; on the other hand, this study has uncovered that the SIL-2R feature holds significant importance in predicting the occurrence of Sarcoidosis, which improved the selection of treatment and its efficacy. A comprehensive understanding of Sarcoidosis is essential to accurately differentiating symptoms associated with this illness from those associated with other conditions. It is strongly recommended that the CatBoost algorithm be used for sarcoidosis prediction.