Computational Data Science

Accomplished 60% accuracy in age prediction, by applying Random Forest for audio and SVR for image data in a multimodal system using facial and vocal cues.
Refined predictive accuracy, by integrating deep learning models (CNNs, MobileNetV2, Wav2Vec2) with late fusion, implemented in PyTorch.

For the final presentation, you can view the PDF here.

For the final report, you can view the PDF here.