Oncogenic Genes Identification Using NLP and Ensemble Learning Approach

Oncogenic gene identification plays a pivotal role in advancing cancer research and treatment strategies. Accurate prediction and classification of genetic mutations are essential for targeted interventions and personalized medicine. This research leverages advanced machine learning techniques, including Naive Bayes, Logistic Regression, Support Vector Machines, Random Forest Classifier, and Partial Least Square Regression, along with ensemble learning, to enhance oncogenic gene detection. The dataset containing the gene variations along with the proper clinical evidence for each variation underwent rigorous pre-processing to ensure uniformity, cleanliness and normalization using one-hot encoding and Word vectorization. Performance metrics such as F1 Score, precision, recall, and accuracy were employed to evaluate the models. The ensemble model emerged as the most effective, as by combining all the machine learning models it achieved an outstanding accuracy rate of 97.8%. This exceptional accuracy signifies a significant advancement over traditional methods and underscores the transformative potential of computational technologies in oncology. The findings pave the way for more precise and personalized cancer treatment regimens.