This is part of my work as a research fellow at the Wharton Social Impact Initiative - WSII. I was part of the Urban Analytics team led by Dr. Shane Jensen, investigating the spatial correlation of schools and crime incidence in Philadelphia. I aggregated all crime, demographic, and environmental variables at the Census Block level. This research involved running a set of 32 linear regressions that combined different school levels, crime types, and time frames. I conducted a propensity score matching analysis to reiterate the results.
This project was done with Eugene Chong for the Land Use and Environmental Modeling course at the University of Pennsylvania. We used data from the 2013 Alberta's Flood to build a logistic model to predict the likelihood of flood for each fishnet cell in Calgary.
In this project, Jane Christen and I modeled the demand for different services offered by Seattle's Aging and Disability Service. We built a set of four logistic regressions to understand the likelihood of elder citizens needing services in the future based on demographics and survey information from previous service users.
This project was done with John Michael Lasalle for the Geospatial Data Analysis in Python course at the University of Pennsylvania. We combined Aqueduct Global Maps 3.0 Data with child mortality rates, nightime light emissions, and population density data to create a cluster analysis of water stress arround the world.
This project is an assignment for the Geospatial Data Analysis in Python course at the University of Pennsylvania. I was interested in exploring different Data Visualization libraries for Python, such as Seaborn, Altair, and Matplotlib.
This is my term project for the course Introduction to Web Mapping at the University of Pennsylvania. I used data from the National Museum of Natural History to create a web application to visualize the Museum’s volcanoes catalog.
This is a Midterm for the course Introduction to Web Mapping at the University of Pennsylvania. I tested some filtering and styling options, using the Seattle 2018 Building Energy Benchmarking dataset.
This is a customized ArcGIS Toolbox that I creating using my experience as a lighting planner. The idea is to streamline a process of feasibility assessment for a LED upgrade plan.
This is a smart city project done for the Sensing the City course at the University of Pennsylvania. I worked with Neil Narayan and Chin Yee Lee to create a prototype for a smart boarding system for SEPTA using sensors.
I have a vast experience in lighting planning and design. Click here to learn more about my planning and design projects.
