Edward Thompson's Linkedin Analytics

I am an accomplished, forward-thinking, and data-focused engineer with an extensive background in providing transformational leadership and serving as a versatile partner within the organization. My skills in the analysis, identification, and remediation of risks has been repeatedly demonstrated in diverse capacities including project leadership, consultancy, and risk management. I have consistently positioned myself as a partner with the organization and clients to drive cohesive and enhanced business practices. Recently completed a Masters of Data Science at UVA. Versed in: Machine Learning (ML), Python, R, Tableau, Statistical Methods, Data Visualization, Root Cause and Risk Analysis. DATA SCIENCE PROJECTS Charlottesville FD Neighborhood Risk Assessment, Language – Python (Capstone Project) June 2021 – Dec 2021 (exp.) Upgrade Neighborhood Risk Assessment to enhance Charlottesville Fire Departments ability to deliver safety programs, improve service delivery, and provide tools to influence city decision and planning. ● Use GIS data expand upon existing framework and expand data to identify program and service improvement metrics. ● Employed Python to evaluate Fire and EMS records using a risk matrix to identify neighborhood level corrective actions to reduce hi expense emergency calls for service. City of Chicago TNC, Language – PySpark, Python (UVA Big Data Systems) Feb 2021 – May 2021 Evaluated effects of City of Chicago’s COVID lockdown on Transportation Network Companies (Uber, Lyft) rides and tipping behaviour of riders based on anonymized ride data. ● Built and tested three distributed computing models to predict a tip event based on ride parameters. ● After lockdown the percentage of remaining riders tipping and the ratio of the tip to the fare decreased. Haiti Relief Project, Language – R (UVA Data Mining) Sep 2020 – Dec 2020 Competed six different ML models to determine best method for directing relief personnel to likely sites of hurricane victims. ● Leveraged Caret package to quickly compete multiple ML models with same dataset. ● Utilized model tuning to achieve multiple models that exceeded 99% accuracy in hold out data. Safety Incident Tracking and Analysis, Language – Excel (Electrolux) Nov 2014 – May 2016 Created a system for tracking data from Consumer Product Safety Commission, Health Canada, and other major retailers. ● Provided increased visibility of issues and capacity for incident reporting to Legal Department. ● Created data driven safety feedback to quality and R&D for continuous product improvement.