Preprocessing

Iterative Self-Training

Evaluate & Visualize t-SNE with Confidence

Autoencoder Training Learn Latent Features

Feature Extraction Using Encoder

Supervised Classification RF on Learned Features

Evaluate & Visualize t-SNE of Latent Space

Combine All Datasets

Load RNA Data (Liver, Lung, Prostate, Thyroid)

Initial Data Exploration

Split Data

Train using Random Forest, XGBoost, LightGBM, SVM, Gradient Boosting etc ML Models

3D Model Performance Plots Accuracy, AUC, Time

Select Top Models Per Cancer

Individual Model Metrics Reports, ROC, F1, Precision, Recall

Create Ensemble Models Voting Classifier

Train & Evaluate Ensembles

Ensemble Model Metrics Comparison Across Cancers

Save Ensemble Models

Final Visualizations: Bar Plots, ROC Curves, PR Curves