Scribd
Upload
64 views9 pages

The Location Recommendatio Ntonew Restaurants: Eva Li

The document discusses factors to consider when selecting a location for a new restaurant, including neighborhood population and demographics, existing competition, and future development opportunities. It presents an analysis of Toronto neighborhoods segmented into 4 clusters based on restaurant data from Foursquare. The analysis shows the distribution and concentration of different restaurant categories across clusters to help investors identify low-competition areas for new restaurants.

Uploaded by

Eva Li
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
64 views9 pages

The Location Recommendatio Ntonew Restaurants: Eva Li

The document discusses factors to consider when selecting a location for a new restaurant, including neighborhood population and demographics, existing competition, and future development opportunities. It presents an analysis of Toronto neighborhoods segmented into 4 clusters based on restaurant data from Foursquare. The analysis shows the distribution and concentration of different restaurant categories across clusters to help investors identify low-competition areas for new restaurants.

Uploaded by

Eva Li
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

THE LOCATION

RECOMMENDATIO
N TO NEW
RESTAURANTS
Eva lI
Introduction
Location is the most important factor of success to open a new restaurant. There are some key
indexes to evaluate if the location is good or not:

HOW'S THE NEIGHBORHOOD IN TERMS OF WHAT'S THE CURRENT COMPETITION WHAT'S THE FUTURE OPPORTUNITY WITHIN
POPULATION TO CHECK IF THE RESTAURANT SITUATION WITHIN THE LOCATION, IN TERMS THE LOCATION, CONSIDERING LOCAL
WOULD MEET LOCAL CONSUMERS' NEEDS OF THE NUMBER OF RESTAURANTS, CUISINES DEVELOPMENT PLAN
OFFERING, COMMENTS, ETC.

This project will conduct an analysis on Toronto neighborhood to set up a model for investors to
make decision on how to select the location for their new restaurants.
Here is the data list used for this analysis:

◦ List of postal codes of Toronto from WIKIPEDIA


Data
◦ Geospatial Coordinates CSV file for Toronto postal Acquisition
codes from http://cocl.us/Geospatial_data
◦ Foursquare API
Data Acquisition
The process of data cleaning on postal
codes of Toronto:
- Read the HTML text using
BeautifulSoup library of the list of
postal codes of Toronto
- Group neighborhood with the same
postal code
- Drop data that without info. of
Borough
Data Acquisition
Merge Geo-spatial coordinates into the
dataframe

- Read the Geospatial Coordinates CSV


file from website:
http://cocl.us/Geospatial_data
- Merge it into Postal Code dataframe
by matching with Postal code column
A map of Toronto by using Folium
library
Explore the
neighborhoods and
segment by using
Foursquare API
By linking with Foursquare API,
we can check out the top100
restaurants in each
neighborhoods within a radius
of 500 metres, with the info. of
restaurant name and category.
And then we grouped them by
neighborhood.
To make the data clear, we
only selected those categories
with ‘restaurant’ in category
name. Here is the distribution of
all types of restaurant at
Toronto.
Data Analysis

I use K-means clustering and


popularity recommendation
filtering aproach to make the
data analysis. With K=4, I split
neighborhood to 4 clusters.

In this stage, I merged cluster


labels to the dataset, and then
grouped the dataset by cluster.
The we got the heatmap of
restaurants in each cluster.
Conclusion
By using the heatmap above, we can
check the frequency of a certain type
of restaurant in each cluster. For
example, Middle Eastern restaurant
frequency is very high in cluster2 than
others, which means if someone want to
start a Middle Eastern restaurant in the
neighborhood of cluster2, the
competition would be fiercer than on
other areas.

You might also like