Portfolio

Evaluated the efficiency of each fire station in the San Francisco Fire Department.
Created a visually compelling Tableau story and dashboard by assessing response times of each station from original dataset of over 550k records.

Participate in the Makeover Monday community project to create more effective visualizations using public data.
Other projects of interest also captured.

Analyzed and visualized individual and aggregated customer consumption data to identify grid inefficiencies such as charging electric vehicles during periods of peak demand.
Data was extracted from individual files and loaded into a Pandas DataFrame. Statistics for three criteria were collected to determine how likely each residence had an electric vehicle.
Goal of identifying and changing behavior through the use of future power company incentives.
Machine Learning Addendum: Exploring classifying meters through grouping with K-means Clustering and K-nearest Neighbor Classification.

Assessed if wind direction and velocity caused a correlated shift to predicted tide levels in Wells, ME using Pandas and Jupyter Notebook.
Extracted data using API and by segmenting requests to avoid limitations.
Plotted data with Seaborn and Matplotlib.
Determined east-northeast and east wind directions have a positive correlation with respect to predicted height. A 20 mph sustained wind from the ENE or E will cause about a 2 ft increase in sea level.

Leveraged Redfin and Foursquare APIs to create a method of recommending homes which fit both client house and neighborhood amenity criteria.
Project returns a list of suitable homes as well as a Folium map of each suitable neighborhood with nearest required amenities shown.