About Me
Hi, this is Kien. I am a Master student, studying Computational Analytics in College of Computing at Georgia Tech. Previously, I have three years working as a data scientist focused on customer analytics and personalization. My research interests include next-best-action systems, recommendation systems, and applications of representation learning and optimization to personalization.
My notable projects includes:
- Lookalike audience system: Built an Auto-ML engine, using gradient boosting tree and mutual information, which would iteratively find and target potential customers to increase click-through rate by ~2.5 times for multiple marketing campaigns.
- Semantic layer about customer: Developed a system of ML models, based on boosted tree, sequence modeling, and multi-task learning, to infer 70 common customer attributes which led to 2 times increase in ML models’ reusability.
- Promotional campaign system: Established a promotion budget allocation system using uplift modeling and integer programming to increase ~90% transaction count and save $35K through a 75% reduction in excessive promotion cost.
- Lifetime value prediction: Applied survival analysis and propensity model to predict churn and Customer lifetime value for non-subscription business
If you want to have a chat, please send me an email (ktran332@gatech.edu or trantrikien239@gmail.com) or reach out via any of the means below:
| Github |
Thank you, and have a nice day!
