Hugo Romero
Back to About

AI Researcher

Universidad Carlos III de Madrid (UC3M) Madrid, Spain 2022/09 – 2023/08
Context

Academic research at UC3M analyzing real customer contracting data for a business partner, with strict constraints on data privacy and no off-the-shelf library that covered the approach needed.

Outcome

Built a custom binary logic algorithm for massive customer data processing and a genetic algorithm system for association rule extraction, delivering results with direct business applications.

Research Focus

Researched evolutionary genetic algorithms for association rule extraction, applied to customer profiling. The aim was to analyze contracting trends and identify cross-sell opportunities for a real business partner.

What I Built

The main contribution was a custom algorithm based on binary logic for processing large volumes of customer data with association rules. There was no existing library that matched the approach, so I built the core logic from scratch.

I also implemented data anonymization techniques to handle the sensitivity of the customer data while keeping the analytical results meaningful.

Key Achievements

  • Created a novel binary logic algorithm for massive customer data processing
  • Delivered research with direct business applications for customer profiling
  • Built ML algorithms from first principles without dedicated libraries
  • Protected sensitive data through comprehensive anonymization techniques