FT-HELC: A Fine-tuned Hybrid Ensemble Technique for Lung Cancer Classification and Prediction Using a Clinical Dataset

Lung cancer continues to be a leading cause of cancer-related deaths worldwide, highlighting the importance of early and accurate detection to ensure better health outcomes for patients. Traditional diagnostic approaches often struggle with high-dimensional data complexity, class imbalance, and computational inefficiency. Machine learning has emerged as a valuable strategy for improving classification accuracy and supporting medical decision-making. In this study, a fine-tuned hybrid ensemble technique called FT-HELC is proposed for lung cancer classification and prediction. The FT-HELC model integrates Support Vector Machine (SVM) and Random Forest (RF) to enhance classification performance. It incorporates the Synthetic Minority Over-sampling Technique (SMOTE) for class balancing and utilizes GridSearchCV for hyperparameter tuning. A publicly available clinical dataset was used and subsequently split into training and validation subsets. The proposed FT-HELC model achieved an accuracy of 95.93% and an AUC score of 97.89%.A comparative analysis with standalone SVM and RF models, as well as recent state-of-the-art techniques, demonstrates that the hybrid FT-HELC approach outperforms them and provides robust and reliable predictions. These results highlight the potential of FT-HELC for early-stage lung cancer detection and its applicability in real-world clinical practice.