Project Name: Predicting Wine Quality
Our project made use of the Wine Quality Dataset from the UCI Machine Learning Repository, which includes 6500 'Vihno Verde' wines from the Portugal region, each with 11 physicochemical measurements and a human-labeled quality score. The goal was to provide value to wine manufacturers and consumers by identifying how certain chemical concentrations influence the average quality of a wine and creating a machine learning model that predicts a wine's quality score. Our best model was a polynomial linear regression model that predicted a wine's quality within 0.7 quality points on average. We also developed an unsupervised K-means clustering model to identify the unique "flavor profiles" of wines. These flavor profiles highlighted the human labelers' bias towards strong (high %ABV) and dry (low sweetness) wines.
If you want someone excited about automation and process improvement!
standardizing the process and making it easily reproducible by others.
creative, mission-focused, and take pride in my work.
project management, operational excellence, IT service management/delivery