gregjasonroberts / ibm-capstone-k-meansclustering Goto Github PK

This file and other associated files constitute my submission to the IBM Applied Data Science Capstone project on Coursera. This project represents my file submission in a 9-part Data Science certification program. Tools used include: Web scraping, Foursquare API, U.S. Census Data API.

The consumer demand for child care, early education, and back-up care has grown in recent years in concert with the growing number of working family households. Many families continue to migrate within the suburbs of large metro areas and given those cities offer proximity to work there's an equally significant demand for surrounding facilities with cost effective services for their children.

A fast growing client is determined to expand to other large metro areas and after internally reviewing the demographics within a segment of the largest cities in the country, have chosen Chicago and Washington, DC for this review stage. Due to the growing populations and high percentage of two-parent household working families, these cities were selected for closer inspection. The data science team is tasked with analyzing the neighborhoods of these large hubs which have shown a successful level of market penetration. It is these neighborhoods that will presume not only a substantial customer base to grow with, but a high likelihood of available labor as well.

The data science team will provide data and insight on the more desirable neighborhoods to target by implementing the following data sets and techniques:

• Part 1: Data summarizing the top 20 U.S. city demographics for labor population and household size.

Data source: The U.S. Census Bureau datasets were included using the Census.gov API. These datasets were used to explore which cities are potentially target rich environments for childcare services and warrant further inspection of its surrounding neighborhoods. The preliminary findings indicated that Chicago and Washington, DC satisfied our clients' requirements for a growing populous and contain deomographics consistent with both parents in the workforce and have children below the age of 6. Further review of the surrounding neighborhoods should yield areas with higher levels of competitor saturation indicating sufficient opportunity for growth and potential labor.

• Part 2: Neighborhood demographics that will indicate favorable environments for deployment.

Data source: After circling Chicago and DC for closer review, web scraping techniques were employed to create datasets from the respective neighborhoods as listed on the associated Wikipedia websites. You can review the detail behind those datasets respectively for Chicago (https://en.wikipedia.org/wiki/List_of_neighborhoods_in_Chicago) and Washington, DC (https://en.wikipedia.org/wiki/Neighborhoods_in_Washington,_D.C.)