Federico Ariton
Data Analyst

A full-stack marketing analytics pipeline simulating a real-world e-commerce business case. Built on Brazil's Olist marketplace dataset — engineered a PostgreSQL database, KPIs, and product segmentation using PCA + KMeans. Delivered as a Streamlit dashboard.

Forecasts 1-, 3-, and 7-day stock prices combining Twitter sentiment analysis with market data. Uses ARIMAX, SARIMAX, LSTM and GRU models on a Lambda architecture with Apache Spark and HDFS.

Compares two AI upscaling models for enhancing retro game visuals. Big data pipeline using Hadoop HDFS and Apache Spark processes DIV2K training data — achieving images up to 10× smaller in file size.

Analyzes production-related factors influencing agricultural export values in Ireland. Employs statistical and predictive modeling using CRISP-DM methodology to uncover trends and provide actionable insights.

Applies clustering techniques to segment e-commerce customers based on behavior, enabling businesses to tailor marketing strategies and improve customer engagement and profitability.

Designed and normalized a relational database schema for employee management to 3NF. Wrote advanced SQL queries using joins, views, and stored procedures for operational analysis.