Machine Learning

Anomaly detection and ground truth labeling of an unlabeled dataset. Dataset used for this project is sourced from the heating, ventilation, and air conditioning (HVAC) system of the Engineering and Science Building at Vanderbilt University.

A customer loan risk prediction model based on bidirectional LSTM neural network that leverages call center data and customer portfolio to refine calling criteria and asset reclamation.

Analysis and Evaluation of three state-of-the-art style tansfer models – StayleBank, GatedGAN, StyleMixer.