Skip to content
/ Edinburgh_Airbnb Public

Perform a spatial join and create an interactive Felt choropleth map of Airbnb density per km2 at data zone granular level.

felt.com/map/Tutorial-Learn-to-Use-Felt-f56pLHJ2RcGUpQON3gtwmC?loc=55.95639,-3.183,13.61z&share=1
1 star 0 forks Branches Tags Activity
Star
Notifications You must be signed in to change notification settings

Stephen137/Edinburgh_Airbnb

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

19 Commits
images
images
 
 
output
output
 
 
README.md
README.md
 
 
edinburgh_airbnb.ipynb
edinburgh_airbnb.ipynb
 
 

Repository files navigation

Inside Edinburgh Airbnb

Inspiration and aim of this project

The Airbnb model, which connects travelers with individuals offering accommodations in their homes or properties, offers several advantages. It provides a wide range of lodging options, often at lower prices than traditional hotels, and allows hosts to earn income from their unused space. It fosters a sense of community and cultural exchange, enabling travelers to immerse themselves in local cultures.

However, this model has its downsides, including concerns about safety and security, potential neighborhood disturbances caused by irresponsible guests, and issues related to housing shortages and the displacement of long-term residents in popular tourist destinations. Additionally, regulatory and tax compliance challenges have been a source of controversy in many regions, and some critics argue that Airbnb can contribute to rising rental costs and reduced housing availability in certain markets.

Inside Airbnb is a mission driven project that provides data and advocacy about Airbnb's impact on residential communities. I reviewed the available data and noted that Edinburgh, Scotland (a city I spent 14 years in) is included which inspired me to perform a data dive. I decided I would try to enrich my analysis by making use of spatial data that I used in another recent project on the Scottish Index of Multiple Deprivation (SIMD).

My overall aim is to create a choropleth map of the area covered by the Edinburgh Airbnb listings to reveal the density of listings in each data zone. I hope to achieve this by deconstructing the problem into the following steps :

  • Step 1: Establish which datazone each of the Airbnb listings falls within by performing a spatial join

  • Step 2: Calculate the number of property listings within each data zone

  • Step 3: Calculate the property listing density per data zone using the calculation above and the area metadata that is included with the data zone shape file

  • Step 4: Plot a choropleth map with appropriate colour ramp to provide insights regarding the density of Airbnb properties across Edinburgh

Import the required packages

import os
import geopandas as gpd
import pandas as pd
import matplotlib.pyplot as plt
%matplotlib inline

Load in our data

Spatial Data

datazones = gpd.read_file('data/shapefiles/sc_dz_11.shp')
print(f"Number of rows: {len(datazones)}")
Number of rows: 6976

datazones.head()
<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; } 
.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
</style>
DataZone Name TotPop2011 ResPop2011 HHCnt2011 StdAreaHa StdAreaKm2 Shape_Leng Shape_Area geometry
0 S01006506 Culter - 01 872 852 424 438.880218 4.388801 11801.872345 4.388802e+06 POLYGON ((-2.27748 57.09527, -2.27644 57.09521...
1 S01006507 Culter - 02 836 836 364 22.349739 0.223498 2900.406362 2.217468e+05 POLYGON ((-2.27354 57.10449, -2.27333 57.10449...
2 S01006508 Culter - 03 643 643 340 27.019476 0.270194 3468.761949 2.701948e+05 POLYGON ((-2.27443 57.10171, -2.27237 57.10046...
3 S01006509 Culter - 04 580 580 274 9.625426 0.096254 1647.461389 9.625426e+04 POLYGON ((-2.26611 57.10133, -2.26599 57.10092...
4 S01006510 Culter - 05 644 577 256 18.007657 0.180076 3026.111412 1.800766e+05 POLYGON ((-2.26013 57.10160, -2.26050 57.10134...

So, as you can our datazones dataset contains 6,976 rows which represent "datazones" as defined and included as part of the Scottish Government Index of Multiple Deprivation (SIMD) and we have some basic metadata associated with these datazones, in particular the area of each zone which we will be using to calculate the density of Airbnb property listings.

type(datazones)
geopandas.geodataframe.GeoDataFrame

Our datazones dataset has been loaded from a shapefile using geopandas and is a GeoDataFrame - we have a geometry column which includes the Polygon geomtery of each of the 6,976 data zones. We can visualize in just one line of code :

datazones.plot()
<AxesSubplot: >

png

Non-spatial data

### create a non-spatial DataFrame from the listings data
edin_listings = pd.read_csv('data/edinburgh_airbnb_listings.csv')
print(f"Number of rows: {len(edin_listings)}")
Number of rows: 8044

edin_listings.head()
<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; } 
.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
</style>
id name host_id host_name neighbourhood_group neighbourhood latitude longitude room_type price minimum_nights number_of_reviews last_review reviews_per_month calculated_host_listings_count availability_365 number_of_reviews_ltm license
0 15420 Rental unit in Edinburgh · ★4.97 · 1 bedroom ·... 60423 Charlotte NaN Old Town, Princes Street and Leith Street 55.957590 -3.188050 Entire home/apt 130 3 511 2023-09-03 3.32 1 16 79 NaN
1 24288 Loft in Edinburgh · ★4.62 · 2 bedrooms · 2 bed... 46498 Gordon NaN Canongate, Southside and Dumbiedykes 55.944983 -3.185293 Entire home/apt 91 3 286 2023-07-24 1.81 1 42 55 NaN
2 48645 Condo in Edinburgh · ★4.86 · 1 bedroom · 2 bed... 221474 Mark NaN Canongate, Southside and Dumbiedykes 55.950720 -3.183050 Private room 129 2 979 2023-09-08 6.37 1 69 68 NaN
3 51505 Rental unit in Edinburgh · ★4.76 · 2 bedrooms ... 236828 Francois NaN New Town West 55.954800 -3.196410 Entire home/apt 170 4 51 2023-08-23 0.33 14 2 32 NaN
4 54188 Boat in Edinburgh · ★4.61 · 3 bedrooms · 2 bed... 253850 Natalie NaN Dalry and Fountainbridge 55.942170 -3.208630 Entire home/apt 225 1 43 2022-07-03 0.32 3 175 0 NaN

So we have 8044 rows which represent Edinburgh Airbnb listings as at 11 September 2023. The data was sourced from Inside Airbnb.

type(edin_listings)
pandas.core.frame.DataFrame

Convert the pandas.DataFrame into a geopandas.GeoDataFrame

Although our Airbnb listings dataset has a longitude and a latitude column this is not enough for it to qualify as a GeoDataFrame. However we can make use of this information to create a geometry column.

We want to create a shapely.geometry.Point for each row, based on the columns longitude and latitude. There are different approaches to this task. One such method is the apply() method of the data frame, together with a lambda function.

from shapely.geometry import Point

# Create a column of empty Point objects
edin_listings['point_geometry'] = [Point() for _ in range(len(edin_listings))]

# Use a lambda function to create Point objects and assign them to the 'point_geometry' column
edin_listings['point_geometry'] = edin_listings.apply(lambda row: Point(row['longitude'], row['latitude']), axis=1)

edin_listings.head()
<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; } 
.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
</style>
id name host_id host_name neighbourhood_group neighbourhood latitude longitude room_type price minimum_nights number_of_reviews last_review reviews_per_month calculated_host_listings_count availability_365 number_of_reviews_ltm license point_geometry
0 15420 Rental unit in Edinburgh · ★4.97 · 1 bedroom ·... 60423 Charlotte NaN Old Town, Princes Street and Leith Street 55.957590 -3.188050 Entire home/apt 130 3 511 2023-09-03 3.32 1 16 79 NaN POINT (-3.18805 55.95759)
1 24288 Loft in Edinburgh · ★4.62 · 2 bedrooms · 2 bed... 46498 Gordon NaN Canongate, Southside and Dumbiedykes 55.944983 -3.185293 Entire home/apt 91 3 286 2023-07-24 1.81 1 42 55 NaN POINT (-3.185293348616568 55.94498320836646)
2 48645 Condo in Edinburgh · ★4.86 · 1 bedroom · 2 bed... 221474 Mark NaN Canongate, Southside and Dumbiedykes 55.950720 -3.183050 Private room 129 2 979 2023-09-08 6.37 1 69 68 NaN POINT (-3.18305 55.95072)
3 51505 Rental unit in Edinburgh · ★4.76 · 2 bedrooms ... 236828 Francois NaN New Town West 55.954800 -3.196410 Entire home/apt 170 4 51 2023-08-23 0.33 14 2 32 NaN POINT (-3.19641 55.9548)
4 54188 Boat in Edinburgh · ★4.61 · 3 bedrooms · 2 bed... 253850 Natalie NaN Dalry and Fountainbridge 55.942170 -3.208630 Entire home/apt 225 1 43 2022-07-03 0.32 3 175 0 NaN POINT (-3.20863 55.94217) 
# explicitly specify the geometry column and the coordinate system
edin_listings = gpd.GeoDataFrame(edin_listings, geometry=edin_listings['point_geometry'], crs="WGS84")
type(edin_listings)
geopandas.geodataframe.GeoDataFrame

edin_listings.crs
<Geographic 2D CRS: EPSG:4326>
Name: WGS 84
Axis Info [ellipsoidal]:
- Lat[north]: Geodetic latitude (degree)
- Lon[east]: Geodetic longitude (degree)
Area of Use:
- name: World.
- bounds: (-180.0, -90.0, 180.0, 90.0)
Datum: World Geodetic System 1984 ensemble
- Ellipsoid: WGS 84
- Prime Meridian: Greenwich

edin_listings
<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; } 
.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
</style>
id name host_id host_name neighbourhood_group neighbourhood latitude longitude room_type price minimum_nights number_of_reviews last_review reviews_per_month calculated_host_listings_count availability_365 number_of_reviews_ltm license point_geometry geometry
0 15420 Rental unit in Edinburgh · ★4.97 · 1 bedroom ·... 60423 Charlotte NaN Old Town, Princes Street and Leith Street 55.957590 -3.188050 Entire home/apt 130 3 511 2023-09-03 3.32 1 16 79 NaN POINT (-3.18805 55.95759) POINT (-3.18805 55.95759)
1 24288 Loft in Edinburgh · ★4.62 · 2 bedrooms · 2 bed... 46498 Gordon NaN Canongate, Southside and Dumbiedykes 55.944983 -3.185293 Entire home/apt 91 3 286 2023-07-24 1.81 1 42 55 NaN POINT (-3.185293348616568 55.94498320836646) POINT (-3.18529 55.94498)
2 48645 Condo in Edinburgh · ★4.86 · 1 bedroom · 2 bed... 221474 Mark NaN Canongate, Southside and Dumbiedykes 55.950720 -3.183050 Private room 129 2 979 2023-09-08 6.37 1 69 68 NaN POINT (-3.18305 55.95072) POINT (-3.18305 55.95072)
3 51505 Rental unit in Edinburgh · ★4.76 · 2 bedrooms ... 236828 Francois NaN New Town West 55.954800 -3.196410 Entire home/apt 170 4 51 2023-08-23 0.33 14 2 32 NaN POINT (-3.19641 55.9548) POINT (-3.19641 55.95480)
4 54188 Boat in Edinburgh · ★4.61 · 3 bedrooms · 2 bed... 253850 Natalie NaN Dalry and Fountainbridge 55.942170 -3.208630 Entire home/apt 225 1 43 2022-07-03 0.32 3 175 0 NaN POINT (-3.20863 55.94217) POINT (-3.20863 55.94217)
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
8039 974513574116367354 Rental unit in Leith · ★New · 1 bedroom · 2 be... 131677538 Conor NaN Western Harbour and Leith Docks 55.986759 -3.188275 Private room 60 2 0 NaN NaN 1 255 0 NaN POINT (-3.1882747601883032 55.98675908383569) POINT (-3.18827 55.98676)
8040 974540296775384595 Rental unit in Edinburgh · ★New · 4 bedrooms ·... 146730114 Adam NaN Newington and Dalkeith Road 55.939714 -3.177628 Entire home/apt 399 2 0 NaN NaN 9 258 0 NaN POINT (-3.1776279 55.9397141) POINT (-3.17763 55.93971)
8041 975083364733576599 Rental unit in Edinburgh · ★New · 3 bedrooms ·... 238856343 Hany NaN Hillside and Calton Hill 55.959688 -3.176968 Entire home/apt 268 2 0 NaN NaN 20 134 0 NaN POINT (-3.176967726675298 55.959687801369945) POINT (-3.17697 55.95969)
8042 976567080851790966 Rental unit in Edinburgh · ★New · 1 bedroom · ... 18970049 Darcy NaN Moredun and Craigour 55.917985 -3.140077 Private room 56 1 0 NaN NaN 2 207 0 NaN POINT (-3.1400765961073613 55.91798460572752) POINT (-3.14008 55.91798)
8043 977248060255868689 Rental unit in Edinburgh · ★New · 2 bedrooms ·... 284574597 Luka NaN Drylaw 55.966904 -3.244546 Entire home/apt 107 1 0 NaN NaN 3 0 0 NaN POINT (-3.244545613737548 55.96690438333705) POINT (-3.24455 55.96690)

8044 rows × 20 columns

Step 1 - Performing a spatial join

We have a Point geometry for each of the property listings. We have a Polygon geometry for each of the data zones. We are now ready to join our GeoDataFrames.

airbnb_datazones = edin_listings.sjoin(datazones, how="left", predicate="within")
airbnb_datazones.head()
<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; } 
.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
</style>
id name host_id host_name neighbourhood_group neighbourhood latitude longitude room_type price ... index_right DataZone Name TotPop2011 ResPop2011 HHCnt2011 StdAreaHa StdAreaKm2 Shape_Leng Shape_Area
0 15420 Rental unit in Edinburgh · ★4.97 · 1 bedroom ·... 60423 Charlotte NaN Old Town, Princes Street and Leith Street 55.957590 -3.188050 Entire home/apt 130 ... 2170 S01008676 Old Town, Princes Street and Leith Street - 03 779 690 338 29.476317 0.294764 4133.369517 294763.155287
1 24288 Loft in Edinburgh · ★4.62 · 2 bedrooms · 2 bed... 46498 Gordon NaN Canongate, Southside and Dumbiedykes 55.944983 -3.185293 Entire home/apt 91 ... 2174 S01008680 Canongate, Southside and Dumbiedykes - 01 631 597 326 5.128574 0.051285 1473.321749 51285.748615
2 48645 Condo in Edinburgh · ★4.86 · 1 bedroom · 2 bed... 221474 Mark NaN Canongate, Southside and Dumbiedykes 55.950720 -3.183050 Private room 129 ... 2179 S01008685 Canongate, Southside and Dumbiedykes - 06 1025 777 489 17.121499 0.171215 2976.316012 171214.990188
3 51505 Rental unit in Edinburgh · ★4.76 · 2 bedrooms ... 236828 Francois NaN New Town West 55.954800 -3.196410 Entire home/apt 170 ... 2343 S01008849 New Town West - 01 727 727 396 20.245548 0.202455 3294.418369 202455.476170
4 54188 Boat in Edinburgh · ★4.61 · 3 bedrooms · 2 bed... 253850 Natalie NaN Dalry and Fountainbridge 55.942170 -3.208630 Entire home/apt 225 ... 2145 S01008651 Dalry and Fountainbridge - 01 965 686 497 16.718419 0.167184 3153.946787 167184.197335

5 rows × 30 columns

Step 2: Calculate the number of property listings within each data zone

We can use .groupby to group the number of properties in each data zone :

# group by DataZone and count # airbnb listings
count_listings = airbnb_datazones.groupby(['DataZone', 'Name', 'StdAreaKm2']).size().reset_index(name='Airbnb_Count')
count_listings.sort_values(by='Airbnb_Count',ascending=False)
<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; } 
.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
</style>
DataZone Name StdAreaKm2 Airbnb_Count
241 S01008677 Old Town, Princes Street and Leith Street - 04 0.438908 218
429 S01008868 Deans Village - 01 0.284072 191
239 S01008675 Old Town, Princes Street and Leith Street - 02 0.200589 159
410 S01008849 New Town West - 01 0.202455 133
240 S01008676 Old Town, Princes Street and Leith Street - 03 0.294764 121
... ... ... ... ...
103 S01008533 Comiston and Swanston - 01 0.137717 1
278 S01008715 Bingham, Magdalene and The Christians - 04 0.264941 1
533 S01008978 Carrick Knowe - 02 0.072332 1
140 S01008575 Liberton East - 03 0.230383 1
111 S01008545 Fairmilehead - 05 0.403233 1

566 rows × 4 columns

number_of_listings = count_listings.Airbnb_Count.sum()
number_of_listings
8044

So our Airbnb listings data covers 566 datazones - sanity check - 566 out of a total of 6976 for the whole of Scotland = 8.1%. Seems reasonable.

# merge the count back into the original GeoDataFrame
airbnb_datazones = airbnb_datazones.merge(count_listings, on='DataZone', how='left')
airbnb_datazones.head()
<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; } 
.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
</style>
id name host_id host_name neighbourhood_group neighbourhood latitude longitude room_type price ... TotPop2011 ResPop2011 HHCnt2011 StdAreaHa StdAreaKm2_x Shape_Leng Shape_Area Name_y StdAreaKm2_y Airbnb_Count
0 15420 Rental unit in Edinburgh · ★4.97 · 1 bedroom ·... 60423 Charlotte NaN Old Town, Princes Street and Leith Street 55.957590 -3.188050 Entire home/apt 130 ... 779 690 338 29.476317 0.294764 4133.369517 294763.155287 Old Town, Princes Street and Leith Street - 03 0.294764 121
1 24288 Loft in Edinburgh · ★4.62 · 2 bedrooms · 2 bed... 46498 Gordon NaN Canongate, Southside and Dumbiedykes 55.944983 -3.185293 Entire home/apt 91 ... 631 597 326 5.128574 0.051285 1473.321749 51285.748615 Canongate, Southside and Dumbiedykes - 01 0.051285 27
2 48645 Condo in Edinburgh · ★4.86 · 1 bedroom · 2 bed... 221474 Mark NaN Canongate, Southside and Dumbiedykes 55.950720 -3.183050 Private room 129 ... 1025 777 489 17.121499 0.171215 2976.316012 171214.990188 Canongate, Southside and Dumbiedykes - 06 0.171215 96
3 51505 Rental unit in Edinburgh · ★4.76 · 2 bedrooms ... 236828 Francois NaN New Town West 55.954800 -3.196410 Entire home/apt 170 ... 727 727 396 20.245548 0.202455 3294.418369 202455.476170 New Town West - 01 0.202455 133
4 54188 Boat in Edinburgh · ★4.61 · 3 bedrooms · 2 bed... 253850 Natalie NaN Dalry and Fountainbridge 55.942170 -3.208630 Entire home/apt 225 ... 965 686 497 16.718419 0.167184 3153.946787 167184.197335 Dalry and Fountainbridge - 01 0.167184 58

5 rows × 33 columns

Step 3 - Calculate the property listing density per data zone

# Step 3: Calculate the density
airbnb_datazones['Density'] = (airbnb_datazones['Airbnb_Count'] / airbnb_datazones['StdAreaKm2_x']).round(1)
airbnb_datazones
<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; } 
.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
</style>
id name host_id host_name neighbourhood_group neighbourhood latitude longitude room_type price ... ResPop2011 HHCnt2011 StdAreaHa StdAreaKm2_x Shape_Leng Shape_Area Name_y StdAreaKm2_y Airbnb_Count Density
0 15420 Rental unit in Edinburgh · ★4.97 · 1 bedroom ·... 60423 Charlotte NaN Old Town, Princes Street and Leith Street 55.957590 -3.188050 Entire home/apt 130 ... 690 338 29.476317 0.294764 4133.369517 294763.155287 Old Town, Princes Street and Leith Street - 03 0.294764 121 410.5
1 24288 Loft in Edinburgh · ★4.62 · 2 bedrooms · 2 bed... 46498 Gordon NaN Canongate, Southside and Dumbiedykes 55.944983 -3.185293 Entire home/apt 91 ... 597 326 5.128574 0.051285 1473.321749 51285.748615 Canongate, Southside and Dumbiedykes - 01 0.051285 27 526.5
2 48645 Condo in Edinburgh · ★4.86 · 1 bedroom · 2 bed... 221474 Mark NaN Canongate, Southside and Dumbiedykes 55.950720 -3.183050 Private room 129 ... 777 489 17.121499 0.171215 2976.316012 171214.990188 Canongate, Southside and Dumbiedykes - 06 0.171215 96 560.7
3 51505 Rental unit in Edinburgh · ★4.76 · 2 bedrooms ... 236828 Francois NaN New Town West 55.954800 -3.196410 Entire home/apt 170 ... 727 396 20.245548 0.202455 3294.418369 202455.476170 New Town West - 01 0.202455 133 656.9
4 54188 Boat in Edinburgh · ★4.61 · 3 bedrooms · 2 bed... 253850 Natalie NaN Dalry and Fountainbridge 55.942170 -3.208630 Entire home/apt 225 ... 686 497 16.718419 0.167184 3153.946787 167184.197335 Dalry and Fountainbridge - 01 0.167184 58 346.9
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
8039 974513574116367354 Rental unit in Leith · ★New · 1 bedroom · 2 be... 131677538 Conor NaN Western Harbour and Leith Docks 55.986759 -3.188275 Private room 60 ... 855 431 36.770579 0.367705 2978.802850 367705.759844 Western Harbour and Leith Docks - 03 0.367705 44 119.7
8040 974540296775384595 Rental unit in Edinburgh · ★New · 4 bedrooms ·... 146730114 Adam NaN Newington and Dalkeith Road 55.939714 -3.177628 Entire home/apt 399 ... 1121 544 5.294937 0.052949 1822.746662 52949.359874 Newington and Dalkeith Road - 05 0.052949 23 434.4
8041 975083364733576599 Rental unit in Edinburgh · ★New · 3 bedrooms ·... 238856343 Hany NaN Hillside and Calton Hill 55.959688 -3.176968 Entire home/apt 268 ... 1019 558 6.646255 0.066463 1711.652482 66462.552191 Hillside and Calton Hill - 01 0.066463 39 586.8
8042 976567080851790966 Rental unit in Edinburgh · ★New · 1 bedroom · ... 18970049 Darcy NaN Moredun and Craigour 55.917985 -3.140077 Private room 56 ... 817 346 20.206043 0.202062 2991.947016 202060.428869 Moredun and Craigour - 04 0.202062 3 14.8
8043 977248060255868689 Rental unit in Edinburgh · ★New · 2 bedrooms ·... 284574597 Luka NaN Drylaw 55.966904 -3.244546 Entire home/apt 107 ... 962 463 16.393302 0.163932 2193.767849 163933.028716 Drylaw - 05 0.163932 4 24.4

8044 rows × 34 columns

So now we have a Density column added to our dataframe, which now gives us of all Airbnb listings in Edinburgh together with the name and geometry of the SIMD datazone that they fall within, and the density of Airbnb properties per datazone.

Calculate overall density index for Edinburgh

number_of_listings
8044

edinburgh_area = count_listings.StdAreaKm2.sum().round(1)
edinburgh_area
245.9

edinburgh_density = (number_of_listings / edinburgh_area).round(1)
edinburgh_density
32.7

So the overall density of Airbnb listings as at 11 September 2023 was 32.7 per km2.

Rank datazones by Airbnb density

Let's drill down and take a look at the density at data zone level.

density_ranking =airbnb_datazones.groupby('DataZone')[['Name_y', 'Density']].max().round(1)
density_ranking = density_ranking.sort_values(by='Density', ascending=False)
density_ranking
<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; } 
.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
</style>
Name_y Density
DataZone
S01008671 Meadows and Southside - 06 1201.6
S01008812 Hillside and Calton Hill - 07 1201.1
S01008662 Tollcross - 04 1147.1
S01008664 Tollcross - 06 1082.1
S01008679 Old Town, Princes Street and Leith Street - 06 1077.0
... ... ...
S01009002 Dalmeny, Kirkliston and Newbridge - 06 0.4
S01008438 Bonaly and The Pentlands - 01 0.4
S01009000 Dalmeny, Kirkliston and Newbridge - 04 0.3
S01008997 Dalmeny, Kirkliston and Newbridge - 01 0.2
S01008418 Balerno and Bonnington Village - 02 0.1

566 rows × 2 columns

We can see that the datazone with the highest density index is Meadows and Southside - 06.

datazone_density = datazones.merge(airbnb_datazones, on='DataZone', how='inner')
datazone_density
<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; } 
.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
</style>
DataZone Name TotPop2011_x ResPop2011_x HHCnt2011_x StdAreaHa_x StdAreaKm2 Shape_Leng_x Shape_Area_x geometry_x ... ResPop2011_y HHCnt2011_y StdAreaHa_y StdAreaKm2_x Shape_Leng_y Shape_Area_y Name_y StdAreaKm2_y Airbnb_Count Density
0 S01008417 Balerno and Bonnington Village - 01 726 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... ... 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 Balerno and Bonnington Village - 01 10.299931 5 0.485440
1 S01008417 Balerno and Bonnington Village - 01 726 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... ... 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 Balerno and Bonnington Village - 01 10.299931 5 0.485440
2 S01008417 Balerno and Bonnington Village - 01 726 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... ... 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 Balerno and Bonnington Village - 01 10.299931 5 0.485440
3 S01008417 Balerno and Bonnington Village - 01 726 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... ... 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 Balerno and Bonnington Village - 01 10.299931 5 0.485440
4 S01008417 Balerno and Bonnington Village - 01 726 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... ... 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 Balerno and Bonnington Village - 01 10.299931 5 0.485440
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
8039 S01009010 Queensferry West - 02 1021 1021 404 20.571212 0.205711 2549.606615 2.057121e+05 POLYGON ((-3.40947 55.98996, -3.40948 55.98955... ... 1021 404 20.571212 0.205711 2549.606615 2.057121e+05 Queensferry West - 02 0.205711 3 14.583566
8040 S01009012 Queensferry West - 04 752 752 296 11.924532 0.119244 1694.365106 1.192453e+05 POLYGON ((-3.41584 55.98721, -3.41533 55.98683... ... 752 296 11.924532 0.119244 1694.365106 1.192453e+05 Queensferry West - 04 0.119244 2 16.772332
8041 S01009012 Queensferry West - 04 752 752 296 11.924532 0.119244 1694.365106 1.192453e+05 POLYGON ((-3.41584 55.98721, -3.41533 55.98683... ... 752 296 11.924532 0.119244 1694.365106 1.192453e+05 Queensferry West - 04 0.119244 2 16.772332
8042 S01009013 Queensferry West - 05 893 893 363 75.940344 0.759404 5408.285433 7.594034e+05 POLYGON ((-3.41600 55.99385, -3.41582 55.99347... ... 893 363 75.940344 0.759404 5408.285433 7.594034e+05 Queensferry West - 05 0.759404 2 2.633644
8043 S01009013 Queensferry West - 05 893 893 363 75.940344 0.759404 5408.285433 7.594034e+05 POLYGON ((-3.41600 55.99385, -3.41582 55.99347... ... 893 363 75.940344 0.759404 5408.285433 7.594034e+05 Queensferry West - 05 0.759404 2 2.633644

8044 rows × 43 columns

datazone_density.columns
Index(['DataZone', 'Name', 'TotPop2011_x', 'ResPop2011_x', 'HHCnt2011_x',
       'StdAreaHa_x', 'StdAreaKm2', 'Shape_Leng_x', 'Shape_Area_x',
       'geometry_x', 'id', 'name', 'host_id', 'host_name',
       'neighbourhood_group', 'neighbourhood', 'latitude', 'longitude',
       'room_type', 'price', 'minimum_nights', 'number_of_reviews',
       'last_review', 'reviews_per_month', 'calculated_host_listings_count',
       'availability_365', 'number_of_reviews_ltm', 'license',
       'point_geometry', 'geometry_y', 'index_right', 'Name_x', 'TotPop2011_y',
       'ResPop2011_y', 'HHCnt2011_y', 'StdAreaHa_y', 'StdAreaKm2_x',
       'Shape_Leng_y', 'Shape_Area_y', 'Name_y', 'StdAreaKm2_y',
       'Airbnb_Count', 'Density'],
      dtype='object')

cols_to_remove = ['point_geometry', 'geometry_y', 'index_right', 'Name_y', 'TotPop2011_y', 'ResPop2011_y', 'HHCnt2011_y', 'StdAreaHa_y', 'StdAreaKm2_y', 'Shape_Leng_y', 'Shape_Area_y']
datazone_density_tidy = datazone_density.drop(columns=cols_to_remove)
datazone_density_tidy
<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; } 
.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
</style>
DataZone Name TotPop2011_x ResPop2011_x HHCnt2011_x StdAreaHa_x StdAreaKm2 Shape_Leng_x Shape_Area_x geometry_x ... last_review reviews_per_month calculated_host_listings_count availability_365 number_of_reviews_ltm license Name_x StdAreaKm2_x Airbnb_Count Density
0 S01008417 Balerno and Bonnington Village - 01 726 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... ... 2023-07-07 0.33 1 237 1 NaN Balerno and Bonnington Village - 01 10.299931 5 0.485440
1 S01008417 Balerno and Bonnington Village - 01 726 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... ... 2023-06-20 0.08 17 241 2 NaN Balerno and Bonnington Village - 01 10.299931 5 0.485440
2 S01008417 Balerno and Bonnington Village - 01 726 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... ... 2023-08-28 0.51 2 257 8 NaN Balerno and Bonnington Village - 01 10.299931 5 0.485440
3 S01008417 Balerno and Bonnington Village - 01 726 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... ... 2023-09-06 1.13 2 272 11 NaN Balerno and Bonnington Village - 01 10.299931 5 0.485440
4 S01008417 Balerno and Bonnington Village - 01 726 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... ... 2021-10-10 0.08 1 327 0 NaN Balerno and Bonnington Village - 01 10.299931 5 0.485440
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
8039 S01009010 Queensferry West - 02 1021 1021 404 20.571212 0.205711 2549.606615 2.057121e+05 POLYGON ((-3.40947 55.98996, -3.40948 55.98955... ... NaN NaN 1 320 0 NaN Queensferry West - 02 0.205711 3 14.583566
8040 S01009012 Queensferry West - 04 752 752 296 11.924532 0.119244 1694.365106 1.192453e+05 POLYGON ((-3.41584 55.98721, -3.41533 55.98683... ... 2023-09-07 0.49 2 0 14 NaN Queensferry West - 04 0.119244 2 16.772332
8041 S01009012 Queensferry West - 04 752 752 296 11.924532 0.119244 1694.365106 1.192453e+05 POLYGON ((-3.41584 55.98721, -3.41533 55.98683... ... 2023-09-02 1.69 2 37 8 NaN Queensferry West - 04 0.119244 2 16.772332
8042 S01009013 Queensferry West - 05 893 893 363 75.940344 0.759404 5408.285433 7.594034e+05 POLYGON ((-3.41600 55.99385, -3.41582 55.99347... ... 2023-04-29 0.15 2 18 1 NaN Queensferry West - 05 0.759404 2 2.633644
8043 S01009013 Queensferry West - 05 893 893 363 75.940344 0.759404 5408.285433 7.594034e+05 POLYGON ((-3.41600 55.99385, -3.41582 55.99347... ... 2023-08-15 0.36 1 73 6 NaN Queensferry West - 05 0.759404 2 2.633644

8044 rows × 32 columns

type(datazone_density_tidy)
pandas.core.frame.DataFrame

datazone_airbnb_density = gpd.GeoDataFrame(datazone_density_tidy, geometry=datazone_density_tidy['geometry_x'], crs="WGS84")

Step 4 - Plot a choropleth map

We now have enough information to plot a choropleth density map.

Now let’s see what it looks like without a classification scheme:

plt.figure(figsize=(20,20))
<Figure size 2000x2000 with 0 Axes>




<Figure size 2000x2000 with 0 Axes>

datazone_airbnb_density.plot(column="Density", cmap="OrRd", edgecolor="k", legend=True)
<AxesSubplot: >

png

Classification by quantiles

Quantiles will create attractive maps that place an equal number of observations in each class: If you have 30 counties and 6 data classes, you’ll have 5 counties in each class. The problem with quantiles is that you can end up with classes that have very different numerical ranges (e.g., 1-4, 4-9, 9-250).

# Splitting the data in four shows some spatial clustering around the center
datazone_airbnb_density.plot(
    column="Density", scheme="quantiles", k=12, cmap="OrRd", edgecolor="k")
<AxesSubplot: >

png

Classificaton by natural breaks

Natural Breaks is a kind of “optimal” classification scheme that finds class breaks that will minimize within-class variance and maximize between-class differences. One drawback of this approach is each dataset generates a unique classification solution, and if you need to make comparison across maps, such as in an atlas or a series (e.g., one map each for 1980, 1990, 2000) you might want to use a single scheme that can be applied across all of the maps.

# Compare this to the previous 3-bin figure with quantiles
datazone_airbnb_density.plot(
    column="Density",
    scheme="natural_breaks",
    k=12,
    cmap="OrRd",
    edgecolor="k",
    legend=False,
)
<AxesSubplot: >

png

These maps are OK for obtaining a quick overview, but they are very crude and static. Let's add some functionality by harnessing Folium.

Interactive Map using Folium

Folium makes it easy to visualize data that’s been manipulated in Python on an interactive leaflet map. It enables both the binding of data to a map for choropleth visualizations as well as passing rich vector/raster/HTML visualizations as markers on the map.

The library has a number of built-in tilesets from OpenStreetMap, Mapbox, and Stamen, and supports custom tilesets. Folium supports both Image, Video, GeoJSON and TopoJSON overlays and has a number of vector layers built-in.

import folium
# Create a Map instance
m = folium.Map(location=[55.9533, -3.1883], zoom_start=10, control_scale=True)
m
Make this Notebook Trusted to load map: File -> Trust Notebook<iframe srcdoc="<!DOCTYPE html> <html> <head> 
&lt;meta http-equiv=&quot;content-type&quot; content=&quot;text/html; charset=UTF-8&quot; /&gt;

    &lt;script&gt;
        L_NO_TOUCH = false;
        L_DISABLE_3D = false;
    &lt;/script&gt;

&lt;style&gt;html, body {width: 100%;height: 100%;margin: 0;padding: 0;}&lt;/style&gt;
&lt;style&gt;#map {position:absolute;top:0;bottom:0;right:0;left:0;}&lt;/style&gt;
&lt;script src=&quot;https://cdn.jsdelivr.net/npm/leaflet@1.9.3/dist/leaflet.js&quot;&gt;&lt;/script&gt;
&lt;script src=&quot;https://code.jquery.com/jquery-1.12.4.min.js&quot;&gt;&lt;/script&gt;
&lt;script src=&quot;https://cdn.jsdelivr.net/npm/bootstrap@5.2.2/dist/js/bootstrap.bundle.min.js&quot;&gt;&lt;/script&gt;
&lt;script src=&quot;https://cdnjs.cloudflare.com/ajax/libs/Leaflet.awesome-markers/2.0.2/leaflet.awesome-markers.js&quot;&gt;&lt;/script&gt;
&lt;link rel=&quot;stylesheet&quot; href=&quot;https://cdn.jsdelivr.net/npm/leaflet@1.9.3/dist/leaflet.css&quot;/&gt;
&lt;link rel=&quot;stylesheet&quot; href=&quot;https://cdn.jsdelivr.net/npm/bootstrap@5.2.2/dist/css/bootstrap.min.css&quot;/&gt;
&lt;link rel=&quot;stylesheet&quot; href=&quot;https://netdna.bootstrapcdn.com/bootstrap/3.0.0/css/bootstrap.min.css&quot;/&gt;
&lt;link rel=&quot;stylesheet&quot; href=&quot;https://cdn.jsdelivr.net/npm/@fortawesome/fontawesome-free@6.2.0/css/all.min.css&quot;/&gt;
&lt;link rel=&quot;stylesheet&quot; href=&quot;https://cdnjs.cloudflare.com/ajax/libs/Leaflet.awesome-markers/2.0.2/leaflet.awesome-markers.css&quot;/&gt;
&lt;link rel=&quot;stylesheet&quot; href=&quot;https://cdn.jsdelivr.net/gh/python-visualization/folium/folium/templates/leaflet.awesome.rotate.min.css&quot;/&gt;

        &lt;meta name=&quot;viewport&quot; content=&quot;width=device-width,
            initial-scale=1.0, maximum-scale=1.0, user-scalable=no&quot; /&gt;
        &lt;style&gt;
            #map_4de7f276fe76bdef71967df546f50ecf {
                position: relative;
                width: 100.0%;
                height: 100.0%;
                left: 0.0%;
                top: 0.0%;
            }
            .leaflet-container { font-size: 1rem; }
        &lt;/style&gt;

</head> <body>

        &lt;div class=&quot;folium-map&quot; id=&quot;map_4de7f276fe76bdef71967df546f50ecf&quot; &gt;&lt;/div&gt;

</body> <script>

        var map_4de7f276fe76bdef71967df546f50ecf = L.map(
            &quot;map_4de7f276fe76bdef71967df546f50ecf&quot;,
            {
                center: [55.9533, -3.1883],
                crs: L.CRS.EPSG3857,
                zoom: 10,
                zoomControl: true,
                preferCanvas: false,
            }
        );
        L.control.scale().addTo(map_4de7f276fe76bdef71967df546f50ecf);





        var tile_layer_29e01124caa938ee2aa12f7d7e85cf4b = L.tileLayer(
            &quot;https://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png&quot;,
            {&quot;attribution&quot;: &quot;Data by \u0026copy; \u003ca target=\&quot;_blank\&quot; href=\&quot;http://openstreetmap.org\&quot;\u003eOpenStreetMap\u003c/a\u003e, under \u003ca target=\&quot;_blank\&quot; href=\&quot;http://www.openstreetmap.org/copyright\&quot;\u003eODbL\u003c/a\u003e.&quot;, &quot;detectRetina&quot;: false, &quot;maxNativeZoom&quot;: 18, &quot;maxZoom&quot;: 18, &quot;minZoom&quot;: 0, &quot;noWrap&quot;: false, &quot;opacity&quot;: 1, &quot;subdomains&quot;: &quot;abc&quot;, &quot;tms&quot;: false}
        ).addTo(map_4de7f276fe76bdef71967df546f50ecf);

</script> </html>" style="position:absolute;width:100%;height:100%;left:0;top:0;border:none !important;" allowfullscreen webkitallowfullscreen mozallowfullscreen></iframe>

Let's now create a layer for our Edinburgh Airbnb listings :

edin_airbnb_listings = pd.read_csv('data/edinburgh_airbnb_listings.csv')
# Create a column of empty Point objects
edin_airbnb_listings['geometry'] = [Point() for _ in range(len(edin_listings))]

# Use a lambda function to create Point objects and assign them to the 'point_geometry' column
edin_airbnb_listings['geometry'] = edin_airbnb_listings.apply(lambda row: Point(row['longitude'], row['latitude']), axis=1)
edin_airbnb_listings = gpd.GeoDataFrame(edin_airbnb_listings, geometry=edin_airbnb_listings['geometry'], crs="WGS84")
edin_airbnb_listings = edin_airbnb_listings[['id', 'neighbourhood', 'latitude', 'longitude', 'geometry']]
edin_airbnb_listings
<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; } 
.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
</style>
id neighbourhood latitude longitude geometry
0 15420 Old Town, Princes Street and Leith Street 55.957590 -3.188050 POINT (-3.18805 55.95759)
1 24288 Canongate, Southside and Dumbiedykes 55.944983 -3.185293 POINT (-3.18529 55.94498)
2 48645 Canongate, Southside and Dumbiedykes 55.950720 -3.183050 POINT (-3.18305 55.95072)
3 51505 New Town West 55.954800 -3.196410 POINT (-3.19641 55.95480)
4 54188 Dalry and Fountainbridge 55.942170 -3.208630 POINT (-3.20863 55.94217)
... ... ... ... ... ...
8039 974513574116367354 Western Harbour and Leith Docks 55.986759 -3.188275 POINT (-3.18827 55.98676)
8040 974540296775384595 Newington and Dalkeith Road 55.939714 -3.177628 POINT (-3.17763 55.93971)
8041 975083364733576599 Hillside and Calton Hill 55.959688 -3.176968 POINT (-3.17697 55.95969)
8042 976567080851790966 Moredun and Craigour 55.917985 -3.140077 POINT (-3.14008 55.91798)
8043 977248060255868689 Drylaw 55.966904 -3.244546 POINT (-3.24455 55.96690)

8044 rows × 5 columns

type(edin_airbnb_listings)
geopandas.geodataframe.GeoDataFrame

Convert to GeoJSON

The input data used for creating Folium maps generally needs to be in GeoJson format. We can convert our shapefiles (stored as GeoDataFrames) to GeoJSON using .features.GeoJson.

edin_airbnb_listings_gjson = folium.features.GeoJson(edin_airbnb_listings, name='Airbnb listings')
# Check the GeoJSON features
edin_airbnb_listings_gjson.data.get('features')[0]
{'id': '0',
 'type': 'Feature',
 'properties': {'id': 15420,
  'neighbourhood': 'Old Town, Princes Street and Leith Street',
  'latitude': 55.95759,
  'longitude': -3.18805},
 'geometry': {'type': 'Point', 'coordinates': [-3.18805, 55.95759]},
 'bbox': [-3.18805, 55.95759, -3.18805, 55.95759]}

Heatmap

Now that we have our Airbnb dataset in GeoJson format let's use Folium's mapping capabilitie.

Folium plugins allow us to use popular plugins available in leaflet. One of these plugins is HeatMap, which creates a heatmap layer from input points.

Let’s visualize a heatmap of the Edinburgh Airbnb listings. HeatMap requires a list of points, or a numpy array as input, so we need to first manipulate the data a bit:

# Get x and y coordinates for each point
edin_airbnb_listings["x"] = edin_airbnb_listings["geometry"].apply(lambda geom: geom.x)
edin_airbnb_listings["y"] = edin_airbnb_listings["geometry"].apply(lambda geom: geom.y)

# Create a list of coordinate pairs
listings = list(zip(edin_airbnb_listings["y"], edin_airbnb_listings["x"]))
from folium.plugins import HeatMap

# Create a Map instance
m = folium.Map(location=[55.953251, -3.188267], tiles='stamentoner', zoom_start=10, control_scale=True)

# Add heatmap to map instance
# Available parameters: HeatMap(data, name=None, min_opacity=0.5, max_zoom=18, max_val=1.0, radius=25, blur=15, gradient=None, overlay=True, control=True, show=True)
HeatMap(listings).add_to(m)

# Alternative syntax:
#m.add_child(HeatMap(points_array, radius=15))

# Show map
m

The red areas show the areas with a high concentration of Airbnb listings.

Save our heatmap to html for sharing

edinburgh_airbnb_heatmap = "output/edinburgh_airbnb_heatmap.html"
m.save(edinburgh_airbnb_heatmap)

Choropleth map

Next, let’s check how we can overlay our datazone or tract map on top of a basemap using Folium’s choropleth method. This method is able to read the geometries and attributes directly from a geodataframe.

# Create a Geo-id which is needed by the Folium (it needs to have a unique identifier for each row)
datazone_density_tidy['geoid'] = datazone_density_tidy.index.astype(str)
datazone_density_tidy
<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; } 
.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
</style>
DataZone Name TotPop2011_x ResPop2011_x HHCnt2011_x StdAreaHa_x StdAreaKm2 Shape_Leng_x Shape_Area_x geometry_x ... reviews_per_month calculated_host_listings_count availability_365 number_of_reviews_ltm license Name_x StdAreaKm2_x Airbnb_Count Density geoid
0 S01008417 Balerno and Bonnington Village - 01 726 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... ... 0.33 1 237 1 NaN Balerno and Bonnington Village - 01 10.299931 5 0.485440 0
1 S01008417 Balerno and Bonnington Village - 01 726 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... ... 0.08 17 241 2 NaN Balerno and Bonnington Village - 01 10.299931 5 0.485440 1
2 S01008417 Balerno and Bonnington Village - 01 726 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... ... 0.51 2 257 8 NaN Balerno and Bonnington Village - 01 10.299931 5 0.485440 2
3 S01008417 Balerno and Bonnington Village - 01 726 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... ... 1.13 2 272 11 NaN Balerno and Bonnington Village - 01 10.299931 5 0.485440 3
4 S01008417 Balerno and Bonnington Village - 01 726 726 299 1029.993186 10.299931 20191.721420 1.029993e+07 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... ... 0.08 1 327 0 NaN Balerno and Bonnington Village - 01 10.299931 5 0.485440 4
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
8039 S01009010 Queensferry West - 02 1021 1021 404 20.571212 0.205711 2549.606615 2.057121e+05 POLYGON ((-3.40947 55.98996, -3.40948 55.98955... ... NaN 1 320 0 NaN Queensferry West - 02 0.205711 3 14.583566 8039
8040 S01009012 Queensferry West - 04 752 752 296 11.924532 0.119244 1694.365106 1.192453e+05 POLYGON ((-3.41584 55.98721, -3.41533 55.98683... ... 0.49 2 0 14 NaN Queensferry West - 04 0.119244 2 16.772332 8040
8041 S01009012 Queensferry West - 04 752 752 296 11.924532 0.119244 1694.365106 1.192453e+05 POLYGON ((-3.41584 55.98721, -3.41533 55.98683... ... 1.69 2 37 8 NaN Queensferry West - 04 0.119244 2 16.772332 8041
8042 S01009013 Queensferry West - 05 893 893 363 75.940344 0.759404 5408.285433 7.594034e+05 POLYGON ((-3.41600 55.99385, -3.41582 55.99347... ... 0.15 2 18 1 NaN Queensferry West - 05 0.759404 2 2.633644 8042
8043 S01009013 Queensferry West - 05 893 893 363 75.940344 0.759404 5408.285433 7.594034e+05 POLYGON ((-3.41600 55.99385, -3.41582 55.99347... ... 0.36 1 73 6 NaN Queensferry West - 05 0.759404 2 2.633644 8043

8044 rows × 33 columns

# Select only needed columns
datazone_density_tidy = datazone_density_tidy[['geoid', 'Density', 'geometry_x', 'DataZone', 'Name_x', 'StdAreaKm2_x', 'Airbnb_Count']]

#check data
datazone_density_tidy.head()
<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; } 
.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
</style>
geoid Density geometry_x DataZone Name_x StdAreaKm2_x Airbnb_Count
0 0 0.48544 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... S01008417 Balerno and Bonnington Village - 01 10.299931 5
1 1 0.48544 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... S01008417 Balerno and Bonnington Village - 01 10.299931 5
2 2 0.48544 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... S01008417 Balerno and Bonnington Village - 01 10.299931 5
3 3 0.48544 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... S01008417 Balerno and Bonnington Village - 01 10.299931 5
4 4 0.48544 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... S01008417 Balerno and Bonnington Village - 01 10.299931 5 
type(datazone_density_tidy)
pandas.core.frame.DataFrame

datazone_density_tidy = gpd.GeoDataFrame(datazone_density_tidy, geometry=datazone_density_tidy['geometry_x'], crs="WGS84")
type(datazone_density_tidy)
geopandas.geodataframe.GeoDataFrame

datazone_density_tidy
<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; } 
.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
</style>
geoid Density geometry_x DataZone Name_x StdAreaKm2_x Airbnb_Count geometry
0 0 0.485440 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... S01008417 Balerno and Bonnington Village - 01 10.299931 5 POLYGON ((-3.35779 55.88156, -3.35752 55.88116...
1 1 0.485440 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... S01008417 Balerno and Bonnington Village - 01 10.299931 5 POLYGON ((-3.35779 55.88156, -3.35752 55.88116...
2 2 0.485440 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... S01008417 Balerno and Bonnington Village - 01 10.299931 5 POLYGON ((-3.35779 55.88156, -3.35752 55.88116...
3 3 0.485440 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... S01008417 Balerno and Bonnington Village - 01 10.299931 5 POLYGON ((-3.35779 55.88156, -3.35752 55.88116...
4 4 0.485440 POLYGON ((-3.35779 55.88156, -3.35752 55.88116... S01008417 Balerno and Bonnington Village - 01 10.299931 5 POLYGON ((-3.35779 55.88156, -3.35752 55.88116...
... ... ... ... ... ... ... ... ...
8039 8039 14.583566 POLYGON ((-3.40947 55.98996, -3.40948 55.98955... S01009010 Queensferry West - 02 0.205711 3 POLYGON ((-3.40947 55.98996, -3.40948 55.98955...
8040 8040 16.772332 POLYGON ((-3.41584 55.98721, -3.41533 55.98683... S01009012 Queensferry West - 04 0.119244 2 POLYGON ((-3.41584 55.98721, -3.41533 55.98683...
8041 8041 16.772332 POLYGON ((-3.41584 55.98721, -3.41533 55.98683... S01009012 Queensferry West - 04 0.119244 2 POLYGON ((-3.41584 55.98721, -3.41533 55.98683...
8042 8042 2.633644 POLYGON ((-3.41600 55.99385, -3.41582 55.99347... S01009013 Queensferry West - 05 0.759404 2 POLYGON ((-3.41600 55.99385, -3.41582 55.99347...
8043 8043 2.633644 POLYGON ((-3.41600 55.99385, -3.41582 55.99347... S01009013 Queensferry West - 05 0.759404 2 POLYGON ((-3.41600 55.99385, -3.41582 55.99347...

8044 rows × 8 columns

# Select only needed columns
datazone_density_tidy.drop(['geometry_x'],axis=1, inplace=True)
# From review of the GeoJSON output earlier the key is 'properties.geoid'
key_on = 'properties.geoid'  # Adjust this path based on your GeoJSON structure
# Create a Map instance
m = folium.Map(location=[55.953251, -3.188267], zoom_start=14, control_scale=True)

# Plot a choropleth map
# Notice: 'geoid' column that we created earlier needs to be assigned always as the first column
folium.Choropleth(
    geo_data=datazone_density_tidy.to_json(),
    name='Edinburgh Airbnb listings as at 11 September 2023',
    data=datazone_density_tidy,
    columns=['geoid', 'Density'],
    key_on=key_on,  # Use the correct key path here
    fill_color='YlOrRd',
    fill_opacity=0.7,
    line_opacity=0.2,
    line_color='white', 
    line_weight=0,
    highlight=False, 
    smooth_factor=1.0,
    threshold_scale=[0, 50, 100, 250, 500, 1000, 1202],
    legend_name= 'Edinburgh Airbnb listings density per datazone as at 11 September 2023 (km2)').add_to(m)

#Show map
m

Display metadata using Tooltips

It is possible to add different kinds of pop-up messages and tooltips to the map. Here, it would be nice to see the name of datazone and the value of the density metreic when you hover the mouse over the map. Unfortunately this functionality is not apparently implemented in the Choropleth method we used before.

We can add tooltips to our map when plotting the polygons as GeoJson objects using the GeoJsonTooltip feature. (following examples from here and here)

For a quick workaround, we plot the polygons on top of the coropleth map as a transparent layer, and add the tooltip to these objects.

Note: this is not an optimal solution as now the polygon geometry get’s stored twice in the output!

# Convert points to GeoJson
folium.features.GeoJson(datazone_density_tidy,  
                        name='Labels',
                        style_function=lambda x: {'color':'transparent','fillColor':'transparent','weight':0},
                        tooltip=folium.features.GeoJsonTooltip(fields=['Name_x','StdAreaKm2_x','Airbnb_Count','Density'],
                                                                aliases = ['Datazone','Area{km2)', '# Airbnb listings', 'Density'],
                                                                labels=True,
                                                                sticky=False
                                                                            )
                       ).add_to(m)

m

Save our choropleth map to html for sharing

edinburgh_airbnb = "output/edinburgh_airbnb_choropleth_map.html"
m.save(edinburgh_airbnb)

Key Takeaways/Insights

This was a very rewarding project. I was able to enrich the granularity of the Inside Airbnb data (which categorized the property listings by 'Neighbourhood') to data zone level, which are a bit like tracts. By carrying out a spatial join I was able to locate which datazones (Polygons) the listings (Points) fell within.

The overall density index for Edinburgh Airbnb listings is 32.7 per km2 (8,044 listings covering an area of 245.9 km2). However, this does not tell the whole story. The top five data zones with the highest density of Airbnb listings are:

  • Meadows and Southside - 06 1201.6
  • Hillside and Calton Hill - 07 1201.1
  • Tollcross - 04 1147.1
  • Tollcross - 06 1082.1
  • Old Town, Princes Street and Leith Street - 06 1077.0

It was also a good opportunity to make use of Folium which is a powerful and versatile, interactive mapping option. In the end, I resorted to using Felt to share my map. I couldn't find a way to share the Folium maps unfortunately. You can see the chorpleth map here.

About

Perform a spatial join and create an interactive Felt choropleth map of Airbnb density per km2 at data zone granular level.

felt.com/map/Tutorial-Learn-to-Use-Felt-f56pLHJ2RcGUpQON3gtwmC?loc=55.95639,-3.183,13.61z&share=1

Resources

Readme
Activity

Stars

1 star

Watchers

1 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages