The battle of the neighborhoods

Introduction

The project is done for the client “Cuppa-Bliss Inc” since they wanted to open a new branch of their coffee shops
in New York city. We were asked to find out the top 5 neighborhoods where they could open a new shop.

Data

The data required is the location of the coffee shops in each neighborhood. To gather the information we used
the neighborhoods information from the JSON file provided and we used four squares to find how many coffee
shops are there.

Methodology

With the data as described above, we did the data preparation first. Along with the JSON file, we used
four squares to find how many coffee shops are there. Then we filtered the neighborhoods with more
no. of Coffee shops, and got a list of neighborhood with only one coffee shop available. Also, the client
mentioned that they would not want to open the coffee shop where there is a StarBucks in the area. We
also excluded the areas with StarBucks on the analysis.

Finally  the k means algorithm was applied to get the clusters of the neighborhoods.

Results

We got the following result :

The map to show the clusters :


Discussion

K-Means algorithm was used because it was a clustering problem.

Conclusion

As per the analysis, we advised the client “Cuppa-Bliss Inc” to open the coffee shop at Kingsbridge
neighborhood in Bronx.

* This report is done as an assignment.