Exploration and cleaning of the data available from DataSF (https://datasf.org/opendata/). Right now the work is focused on evictions, property taxes, and police data. I mine interesting results such as (1) the clustering of crime reports by type of crie and day of week, (2) distributions of property tax exemptions and values of assessed features (land, improvement, fixtures, ...), and (3) eviction attempts by neighborhood and year (Lakeshore is the new epicenter of evictions!).
The project ambition is to eventually combine the datasets and build interesting prediction tools and interactive dashboards. I am particularly interested in predicting property values to as it intersects with many domains that excite me (e.g. finance, real estate, time-series, geographical data).
At present the project consists of a set of jupyter notebooks. These notebooks use standard data science packages. I recommend installing the needed packages using Anaconda.
Packages include: pandas, numpy, matplotlib, seaborn.
- David Erickson - Initial work - github
See also the list of contributors who participated in this project.
This project is licensed under the MIT License - see the LICENSE.md file for details
- Big thanks to [Data SF] (https://datasf.org/opendata/) for providing this data.