A Comparative Study of Random Forest and Gradient Boosting Machine Learning Algorithms for Tumor Classification in Biomedical Images

This examination leads a relative investigation of Random Forest (RF) and Gradient Boosting Machine (GBM) calculations for cancer grouping in biomedical pictures. Using openly accessible datasets enveloping assorted cancer types and imaging modalities, it assesses the performance of RF and GBM concerning arrangement exactness, accuracy, review, and F1 score. Results show that GBM reliably outperforms RF across various datasets, accomplishing higher characterization exactness and better harmony among accuracy and review. Moreover, it examines the computational productivity of the two calculations, with RF for the most part requiring less preparation time contrasted with GBM. In any case, the predominant performance of GBM legitimizes its more extended preparing times, featuring its adequacy in discriminative undertakings like cancer arrangement. This study adds to the developing group of writing on machine learning applications in clinical imaging examination, underscoring the significance of calculation choice and boundary tuning in creating precise and dependable demonstrative apparatuses. Future exploration headings might incorporate investigating the interpretability and vigor of GBM models in true clinical settings, as well as researching group learning procedures to additionally further develop growth order performance.