Advancing Oropharyngeal Cancer Prognosis: A Novel Ensemble Machine Learning Approach

Authors : Pınar Karadayi Ataş

Pages : 24-40

View : 42 | Download : 39

Publication Date : 2023-12-21

Article Type : Research

Abstract :The use of machine learning algorithms to forecast survival rates in patients with oropharyngeal cancer is the main focus of this study. Given the complexity and variability inherent in cancer prognosis, traditional predictive models often fall short in accuracy and reliability. We used a variety of machine learning methods, each with their own advantages in data analysis, to tackle these problems, including Gaussian Naive Bayes, Random Forest, Gradient Boosting, Linear Support Vector Machine, Logistic Regression, and K-Nearest Neighbors. The development of an ensemble model that combined these separate algorithms was the key to our strategy. The overall predictive power of this model is increased by utilizing the combined advantages of all the techniques. The results of our comparative analysis indicated that although the performance of the individual algorithms varied, the suggested ensemble model performed better than all of them, obtaining higher accuracy, f1-score, precision, and recall. The study\'s findings highlight the potential of ensemble machine learning models in the complex field of cancer prognosis in particular, for medical diagnostics. The ensemble model offers a more comprehensive tool for predicting survival outcomes in patients with oropharyngeal cancer by efficiently combining multiple algorithms. This method not only increases the predictive accuracy but also provides a deeper comprehension of the dynamics of the disease, opening the door to more individualized and successful treatment plans.
Keywords : Veri Bilimi, Orofaringeal Kanser, Tıbbi Teşhis, Makine Öğrenmesi