Wenjin Z.

Data Scientist - Ai Ml Modeling and Application @Best Buy

Boston, MA, US

Developed and deployed advanced deep generative models - Variational Autoencoder (VAE) with transfer learning to synthesize sociodemographic and household–individual data, which has been used as a foundation framework for further applications.• Co-developed and implemented a fused deep learning framework that combined deep learning and predictive model for business strategy research, improved the capability to detect market trends and competitive impacts, customer behavior shifts, and emerging opportunities.• Developed and fine-tuned a BERT-based natural language processing (NLP) model for a reversed supply chain research initiative, generated actionable insights to identify product quality issues, customer experience, and operational efficiency.• Delivered actionable insights from deep learning model outputs, leveraged with Explainable AI (XAI) techniques (e.g, SHAP values) to interpret feature importance and model behavior, enabled business teams to understand the opportunities of growth and optimized resource allocation.