Carson Lloyd Email & Phone Number

Austin, Texas Metropolitan Area

Austin, Texas, United States

Fort Worth, Tx, United States

Worked intensively in Python, Statistical Analysis, Machine Learning, and Deep Learning in this 15-week training program. As part of the program I attended lectures, completed individual and group programming assignments, code challenges and projects to demonstrate a comprehensive understanding of both the theory and applications necessary to become a Data Science professional.• Studied a variety of data science tools in depth, including Python, SQL, scikit-learn, Pandas, and Tableau• Worked both individually and in teams to develop and present a variety of data science projects including:Horse Racing in Hong Kong• Designed an ensemble of Decision Tree models to predict the outcome of horse races.• Obtained data of over 4,000 races by web scraping the Hong Kong Jockey Club website• Developed new features from the data that provided the model with it’s most insightful information based on precision and recall metrics.• Strategically chose a confidence threshold that resulted in the model predicting the top two horses in races with 99% accuracy when meeting the threshold.Tanzania Water Well Crisis• Developed a Gradient Boosted Decision Tree model to classify the functionality of water wells in Tanzania.• Utilized confusion matrices to tune model to minimize false positive rate to decrease waste in resources and increase overall usefulness in real world use cases.• Calculated optimal hyperparameters with hyperopt.• Achieved rank of top 8% in competition on drivendata.orgFinding Flips• Designed a generalized additive model to predict housing prices in King County, Washington.• Created additional dataset by web scraping current house listings in King County on Zillow.com• Performed in-depth data analysis to transform the data to a Gaussian distribution in order to fit the assumptions of a generalized linear model.• Developed tensors for the model in order to incorporate the feature interactions found in the data.