Visualizing Hotels in Senegal with Python

In this project, we will try to extract data from google earth pro in kml format, then convert them to xlsx file (excel format). We only select the necessary columns such as name of hotels, addresses, tel and coordinates. Our file will not include geometry column at the moment, later on we will see how to convert the excel file or csv file into geojson or shapefile then visualize our locations.
Extracting data (scrapping) from google earth pro
When you open a google earth pro software which is totally free now, on the search bar, enter your search, it might appear many pages, for example in my case of study today, i've entered hotels in Senegal, there are a lot of pages, just go to next panel (an icon like a Folder telling you 'copy the current research to a file'), click on that icon, then scroll down and check it, you will see the same title you've entered on the research panel, right click to rename it, for example, give it a name 'page1 hotels in Senegal', then right click on it and select 'save place as ', choose 'kml'. '
Next: You need to convert the kml file format to xlsx file or csv file.
Go to file where you saved your kml file, then right click on it, select 'properties', in 'open with', select 'change', then change it to notepad or note depends on your computer.
Next, open an empty excel file, select 'file', select 'open', find your file and open it, some panels wiill appear just select the first ones always, you can change the column names and choose the interested columns, the rest you can delete it and save your file as xlsx or csv file.
Loading the necessary libraries and file
In [2]:
import pandas as pd
import geopandas
import matplotlib.pyplot as plt
import contextily as ctx 
import os
import numpy as np # you might not need it at the moment, just iis one of the best libraries for data analysis
In [3]:
# Loading my file 
sen_hotels = pd.read_excel('data/some_hotels_in_Senegal.xlsx')
In [4]:
sen_hotels.head()
Out[4]:
hotel_names addresses phone_number coordinates
0 Radisson Blu Hotel, Dakar Sea Plaza Route de la Corniche O, Dakar 16868, Senegal +221 33 869 33 33 -17.4730893,14.6945826,0
1 Radisson Hotel Dakar Diamniadio Prolongement Autoroute a peage a cote du centr... +221 32 824 48 48 -17.1970069,14.7385608,0
2 Paradialaw Senegal, Toubab Dialao, toubab dialaw , prés d... NaN -17.14559,14.5963946,0
3 Hotel Club Royal Saly Route Saly-Ngaparou M'bour SN, 23002, Senegal +221 33 939 52 30 -17.0221955,14.4432474,0
4 Le Grand Hôtel de Thiés Thies, Senegal +221 33 951 00 98 -16.9280314,14.7950366,0
In [5]:
# Checking the coclumns of my data
sen_hotels.columns.values # i also can use sen_hotels.columns
Out[5]:
array(['hotel_names', 'addresses', 'phone_number', 'coordinates'],
      dtype=object)
In [6]:
# Checking the description of my data
sen_hotels.describe()
Out[6]:
hotel_names addresses phone_number coordinates
count 30 30 28 30
unique 30 30 28 30
top ibis Dakar Avenue Nelson Mandela, Dakar, Senegal +221 33 869 33 33 -17.0221955,14.4432474,0
freq 1 1 1 1
We can see there are two missing values in phone_number column, if we are really interested on, like doing a marketing business, we can google the name of the hotels, find the number and complete our data. But here, we are in a case of study, let's see which hotels are missing with phone number
In [7]:
null_data = sen_hotels[sen_hotels.isnull().any(axis=1)]
In [8]:
null_data
Out[8]:
hotel_names addresses phone_number coordinates
2 Paradialaw Senegal, Toubab Dialao, toubab dialaw , prés d... NaN -17.14559,14.5963946,0
12 Mbind Siga Mbind Siga, Mbodiene, Senegal NaN -16.924339,14.8047,0
The most two populated regions in Senegal (which count 14 regions) are 'Dakar' and 'Thies'. These two regions are next to each other and it's normal, because Dakar is the Capital City of Senegal and Thies is the closest region of Dakar. Let's do some text mining with our data, we want to extract only those hotels in Ddakar, for that we make sure that 'Dakar' is in the addresses.
In [9]:
# Extracting hotels in Dakar with our data
dakar_hotels = sen_hotels[sen_hotels['addresses'].str.contains('Dakar')]
In [10]:
dakar_hotels
Out[10]:
hotel_names addresses phone_number coordinates
0 Radisson Blu Hotel, Dakar Sea Plaza Route de la Corniche O, Dakar 16868, Senegal +221 33 869 33 33 -17.4730893,14.6945826,0
5 Boutique Hôtel La Villa Racine 37, Senegal extension Immeuble Kébé, Rue Jules... +221 33 889 41 41 -17.436217,14.6663435,0
7 ibis Dakar Avenue Abdoulaye Fadiga, Dakar 18524, Senegal +221 33 829 59 59 -17.4268162,14.6687365,0
9 Terrou-Bi Resort Boulevard Martin Luther King Dakar, 11500, Sen... +221 33 839 90 39 -17.4660981,14.6769092,0
10 Le Ndiambour - Hôtel et Résidence Dakar Senegal SN, 121 Rue Carnot, Dakar 11000, Senegal +221 33 889 42 89 -17.4395282,14.6674174,0
15 International VDN Hotel Dakar cices foire 651 Dakar 38233, Senegal +221 78 436 91 17 -17.4677832,14.7452296,0
16 La Madrague Ngor Traversée vers l'ile de Ngor(2eme Plage), Daka... +221 33 820 02 23 -17.5104885,14.7500204,0
17 Le Djoloff Dakar, Senegal +221 33 889 36 30 -17.462291,14.679639,0
19 Pullman Dakar Teranga Place de l'Independance, 10 Rue PL 29, Dakar 1... +221 33 889 22 00 -17.4307642,14.667648,0
20 Le Lodge des Almadies SN, Route des Almadies, Dakar 29339, Senegal +221 33 869 03 45 -17.5150118,14.7398163,0
21 Hotel Sokhamon Avenue Nelson Mandela, Dakar, Senegal +221 33 889 71 00 -17.441901,14.663395,0
22 Hotel Jardin Savana Dakar Corniche Est Dakar Plateau, Dakar, Senegal +221 33 849 42 42 -17.4315413,14.6534857,0
23 Hotel Lagon Rte de la Corniche Estate, Dakar, Senegal +221 33 889 25 25 -17.4273289,14.667817,0
25 Hotel Cabourg SN, Dakar 12000, Senegal +221 33 860 07 04 -17.491721,14.7546323,0
28 La Demeure Route de la Corniche O, Dakar, Senegal +221 33 820 76 79 -17.507246,14.735484,0
The above results are all hotels located in Dakar
We can use the same code to extract hotels in 'Thies'
In [11]:
# Extracting hotels in  Thies with our data
thies_hotels = sen_hotels[sen_hotels['addresses'].str.contains('Thies')]
In [12]:
thies_hotels
Out[12]:
hotel_names addresses phone_number coordinates
4 Le Grand Hôtel de Thiés Thies, Senegal +221 33 951 00 98 -16.9280314,14.7950366,0
We have 1 hotel located in Thies in our data
Let's check how many hotels we have in our data
In [13]:
len(sen_hotels)
Out[13]:
30
In [14]:
# How many hotels in Dakar
len(dakar_hotels)
Out[14]:
15
What about the rest of hotels location in senegalese regions
In [15]:
len(sen_hotels) - len(dakar_hotels) - len(thies_hotels)
Out[15]:
14
There are 14 hotels which are not in Dakar and not in Thies
When i opened my data in excel file, i noticed in column of addresses, there are locations like "M'bour, Sally", those are in the region of Thies. In other side, there are locations like 'Toubab Diallao and so on' are located in Dakar, but those information i htink someone who visited Senegal or know Senegal will know that distinction.
Let's move on to the second party of our data analysis today, visualizing the addresses
Senegal and Geospatial
The title Senegal and Geospatial can be strange, but the meaning is to play some geospatial function over hotels in Senegal
There are two spatial data that we can use: Vector and Rasta\ Vector data is when we record data as point, line, polygon and so on.\ Rasta is mostly a satelite image ....
Exploring the hotels in Senegal
let's check the coordinates column in our data to do some processing
Our data contains the coordinates of the point locations of hotels, we would like our data contains the type of hotels (1 star to 5 stars), unfortunetelly, we don't have them. Let's create two columns (x and y), x will represent t he latitude and y the longitude.
In [16]:
sen_hotels['coordinates'].head()
Out[16]:
0    -17.4730893,14.6945826,0
1    -17.1970069,14.7385608,0
2      -17.14559,14.5963946,0
3    -17.0221955,14.4432474,0
4    -16.9280314,14.7950366,0
Name: coordinates, dtype: object
The numbers starting with - are the longitude and numbers starting with 14 are the latitude. We need to remove ,0 in our data
In [17]:
sen_hotels['coordinates'] = sen_hotels['coordinates'].str.replace(",0", "")
In [18]:
# check our data
sen_hotels.head()
Out[18]:
hotel_names addresses phone_number coordinates
0 Radisson Blu Hotel, Dakar Sea Plaza Route de la Corniche O, Dakar 16868, Senegal +221 33 869 33 33 -17.4730893,14.6945826
1 Radisson Hotel Dakar Diamniadio Prolongement Autoroute a peage a cote du centr... +221 32 824 48 48 -17.1970069,14.7385608
2 Paradialaw Senegal, Toubab Dialao, toubab dialaw , prés d... NaN -17.14559,14.5963946
3 Hotel Club Royal Saly Route Saly-Ngaparou M'bour SN, 23002, Senegal +221 33 939 52 30 -17.0221955,14.4432474
4 Le Grand Hôtel de Thiés Thies, Senegal +221 33 951 00 98 -16.9280314,14.7950366
In [19]:
# Split coordinates columns into two columns x and y
sen_hotels[['x','y']] = sen_hotels.coordinates.str.split(",",expand=True,)
In [20]:
sen_hotels.head()
Out[20]:
hotel_names addresses phone_number coordinates x y
0 Radisson Blu Hotel, Dakar Sea Plaza Route de la Corniche O, Dakar 16868, Senegal +221 33 869 33 33 -17.4730893,14.6945826 -17.4730893 14.6945826
1 Radisson Hotel Dakar Diamniadio Prolongement Autoroute a peage a cote du centr... +221 32 824 48 48 -17.1970069,14.7385608 -17.1970069 14.7385608
2 Paradialaw Senegal, Toubab Dialao, toubab dialaw , prés d... NaN -17.14559,14.5963946 -17.14559 14.5963946
3 Hotel Club Royal Saly Route Saly-Ngaparou M'bour SN, 23002, Senegal +221 33 939 52 30 -17.0221955,14.4432474 -17.0221955 14.4432474
4 Le Grand Hôtel de Thiés Thies, Senegal +221 33 951 00 98 -16.9280314,14.7950366 -16.9280314 14.7950366
x is the longitude and y is the latitude
In [21]:
# Check our data type
sen_hotels.dtypes
Out[21]:
hotel_names     object
addresses       object
phone_number    object
coordinates     object
x               object
y               object
dtype: object
This is not normal, because the corrdinates column is numeric and the x and y also, let's first try to remove the white space before going to another solution
In [22]:
#Remove white space at both ends:
sen_hotels.coordinates = sen_hotels.coordinates.str.strip()
sen_hotels.x = sen_hotels.x.str.strip()
sen_hotels.y = sen_hotels.y.str.strip()
In [23]:
sen_hotels.dtypes
Out[23]:
hotel_names     object
addresses       object
phone_number    object
coordinates     object
x               object
y               object
dtype: object
In [24]:
sen_hotels[['x', 'y']] = sen_hotels[['x', 'y']].apply(pd.to_numeric)
In [25]:
sen_hotels.dtypes
Out[25]:
hotel_names      object
addresses        object
phone_number     object
coordinates      object
x               float64
y               float64
dtype: object
Now, i will make a copy of my data to keep it for future error or any unexpected error cases
In [26]:
data = sen_hotels.copy()
In [27]:
# Make a plot of all points
fig, ax = plt.subplots()
ax.plot(sen_hotels.x, sen_hotels.y, 'o')
plt.show()
We have in the above figure a different locations of the hotels depends on the coordinates, but we need to see them in a map, let's add a background map using contextily library
The image is not clear, this is due to the crs (coordinate reference system), we still did not convert our xlsx format to geojson or spatial format., let's do that first
In [28]:
# Converting xlsx to geospatial processing 
spatial = geopandas.GeoDataFrame(
    sen_hotels, geometry=geopandas.points_from_xy(sen_hotels.x, sen_hotels.y))
In [29]:
spatial.head()
Out[29]:
hotel_names addresses phone_number coordinates x y geometry
0 Radisson Blu Hotel, Dakar Sea Plaza Route de la Corniche O, Dakar 16868, Senegal +221 33 869 33 33 -17.4730893,14.6945826 -17.473089 14.694583 POINT (-17.47309 14.69458)
1 Radisson Hotel Dakar Diamniadio Prolongement Autoroute a peage a cote du centr... +221 32 824 48 48 -17.1970069,14.7385608 -17.197007 14.738561 POINT (-17.19701 14.73856)
2 Paradialaw Senegal, Toubab Dialao, toubab dialaw , prés d... NaN -17.14559,14.5963946 -17.145590 14.596395 POINT (-17.14559 14.59639)
3 Hotel Club Royal Saly Route Saly-Ngaparou M'bour SN, 23002, Senegal +221 33 939 52 30 -17.0221955,14.4432474 -17.022195 14.443247 POINT (-17.02220 14.44325)
4 Le Grand Hôtel de Thiés Thies, Senegal +221 33 951 00 98 -16.9280314,14.7950366 -16.928031 14.795037 POINT (-16.92803 14.79504)
We have a new column 'geometry', we see the POINT representing the vector point location for every hotel location address.
In [30]:
spatial.columns
Out[30]:
Index(['hotel_names', 'addresses', 'phone_number', 'coordinates', 'x', 'y',
       'geometry'],
      dtype='object')
In [31]:
spatial.to_file('D:\Research\PROJECT\SENEGAL DATA ANALYSIS\data\spatial_data\hotel_thies_dakar')
In [32]:
# saving to geojson
spatial.to_file('D:\Research\PROJECT\SENEGAL DATA ANALYSIS\data\spatial_data\hotel_thies_dakar.geojson')
In [33]:
spatial.to_excel('D:\Research\PROJECT\SENEGAL DATA ANALYSIS\data\spatial_data\hotels_thies_dakar.xlsx')
let's plot our spatial data
In [34]:
spatial.plot()
Out[34]:
<AxesSubplot:>
We have almost the same figure as sen_hotels.plot() above
In [35]:
# Checking our coordinates reference system
spatial.crs
In [36]:
print(spatial.crs)
None
In [37]:
spatial.crs = 'EPSG:4201'
In [38]:
spatial.crs
Out[38]:
<Geographic 2D CRS: EPSG:4201>
Name: Adindan
Axis Info [ellipsoidal]:
- Lat[north]: Geodetic latitude (degree)
- Lon[east]: Geodetic longitude (degree)
Area of Use:
- name: Eritrea; Ethiopia; South Sudan; Sudan.
- bounds: (21.82, 3.4, 47.99, 22.24)
Datum: Adindan
- Ellipsoid: Clarke 1880 (RGS)
- Prime Meridian: Greenwich
In [39]:
# plot Linz
ax = spatial.plot(figsize=(10, 10), zorder=10, ec='gray', alpha=0.4)

# plot basemap (it uses 'epsg:3857')
src_basemap = ctx.providers.Stamen.Terrain
ctx.add_basemap(ax, alpha=0.6)
d:\programm files\python 3 8 6\lib\site-packages\contextily\tile.py:632: UserWarning: The inferred zoom level of 28 is not valid for the current tile provider (valid zooms: 0 - 18).
  warnings.warn(msg)
Let's Explore more on our spatial data
In [40]:
# Check what kind of object districts is
print(type(spatial))
 
# Check the type of the geometry attribute
print(type(spatial.geometry))
 
<class 'geopandas.geodataframe.GeoDataFrame'>
<class 'geopandas.geoseries.GeoSeries'>
We have two main regions in Senegal in this data, Dakar and Thies, let's extract those two regions from OpenstreetMap and place our hotels on them
In [41]:
## Exploring Dakar and Thies with osmnx, folium, leaf
Explore with GeoPanDas
In [42]:
# get the geometries from my DataFrame (or GeoDataFrame)
gp = spatial.loc[0,'geometry'] # Polygon
gm = spatial.loc[1,'geometry'] # Multi-Polygon
In [43]:
gp
Out[43]:
In [44]:
gm
Out[44]:
In [ ]:
 
In [45]:
import osmnx as ox
import networkx as nx
In [46]:
#dakar_region = ox.project_graph(ox.graph_from_place('Dakar, Senegal', network_type='drive'))
#dd = ox.consolidate_intersections(dakar_region, tolerance=10, rebuild_graph=True, dead_ends=True)
In [47]:
dakar = ox.geocode_to_gdf('Dakar, Senegal', which_result=1)
In [48]:
dakar
Out[48]:
geometry place_name bbox_north bbox_south bbox_east bbox_west
0 POINT (-17.44794 14.69342) Dakar, 2084, Senegal 14.853425 14.533425 -17.287938 -17.607938
In [49]:
dakar.crs
Out[49]:
<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
- Ellipsoid: WGS 84
- Prime Meridian: Greenwich
In [50]:
# Get a graph of Dakar
dk_graph = ox.graph_from_place('Dakar, Senegal')
In [51]:
dk_graph
Out[51]:
<networkx.classes.multidigraph.MultiDiGraph at 0x4c36f700>
In [52]:
ox.plot_graph(dk_graph)
Out[52]:
(<Figure size 576x576 with 1 Axes>, <AxesSubplot:>)
In [53]:
# Extracting 'Thies' 
thies_shape_region = ox.graph_from_place('Thies, Senegal')
ox.plot_graph(thies_shape_region)
Out[53]:
(<Figure size 576x576 with 1 Axes>, <AxesSubplot:>)
Now, we have 'Dakar' and 'Thies' shapes, let's save it
In [65]:
# ox.save_graph_shapefile('D:\Research\PROJECT\SENEGAL DATA ANALYSIS\data\sthies_shape.shp')
# ox.save_graph_xml(thies_shape_region, 'thies_shape_file.shp')
ox.save_graph_shapefile(thies_shape_region)
We can play around and find the network of 'Thies region'
In [59]:
thies_region = ox.graph_from_address('Thies, Senegal')
In [55]:
ox.plot_graph(thies_region)
Out[55]:
(<Figure size 576x576 with 1 Axes>, <AxesSubplot:>)
Let's also extract geopandas data for 'Thies' as we early did with 'Dakar'
In [67]:
import folium
dakar_folium = folium.Map(location=[45.5236, -122.6750])
point_thies = [14.790877871547798, -16.93448289957603]
thies_folium = folium.Map(location=[14.790877871547798, -16.93448289957603])
thies_folium
Out[67]:
Make this Notebook Trusted to load map: File -> Trust Notebook
In [70]:
# Creating a map of Senegal and locating both 'Dakar' and 'Thies'
world = geopandas.read_file(geopandas.datasets.get_path('naturalearth_lowres'))
In [71]:
world.plot()
Out[71]:
<AxesSubplot:>
In [72]:
# Let's find some data in world
world.head()
Out[72]:
pop_est continent name iso_a3 gdp_md_est geometry
0 920938 Oceania Fiji FJI 8374.0 MULTIPOLYGON (((180.00000 -16.06713, 180.00000...
1 53950935 Africa Tanzania TZA 150600.0 POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...
2 603253 Africa W. Sahara ESH 906.5 POLYGON ((-8.66559 27.65643, -8.66512 27.58948...
3 35623680 North America Canada CAN 1674000.0 MULTIPOLYGON (((-122.84000 49.00000, -122.9742...
4 326625791 North America United States of America USA 18560000.0 MULTIPOLYGON (((-122.84000 49.00000, -120.0000...
In [73]:
world.columns
Out[73]:
Index(['pop_est', 'continent', 'name', 'iso_a3', 'gdp_md_est', 'geometry'], dtype='object')
In [74]:
# Let's extract Senegal 
senegal = world[world['name'] == 'Senegal']
In [75]:
senegal
Out[75]:
pop_est continent name iso_a3 gdp_md_est geometry
51 14668522 Africa Senegal SEN 39720.0 POLYGON ((-16.71373 13.59496, -17.12611 14.373...
In [76]:
senegal.plot()
Out[76]:
<AxesSubplot:>
In [80]:
# We have Senegal shape now, let's map the hotels on it
# Make a multi-layered plot
fig, ax = plt.subplots(figsize=(10, 10))
senegal.plot(ax=ax, color='grey')
spatial.plot(ax=ax, color='red', markersize=2)
# Remove the box, ticks and labels
ax.set_axis_off()
plt.show()
We visualized the hotels in Senegal, they are all in the western side of the country. Both 'Thies' and 'Dakar' are located in the same area.
Conclusion
Analyzing spatial data is a fun, we can create our own real data from google earth, tranform the data to spatial data, visualize it and get a nice result.
Search This Blog

My Experiences in Python Programming Language

Visualizing hotels in Senegal (two regions Dakar and Thies)

Visualizing Hotels in Senegal with Python

Extracting data (scrapping) from google earth pro

Loading the necessary libraries and file

Senegal and Geospatial

Exploring the hotels in Senegal

Let's Explore more on our spatial data

Explore with GeoPanDas

Comments

Post a Comment

Popular posts from this blog

Farming land size owned by households per region in Senegal

	hotel_names	addresses	phone_number	coordinates
0	Radisson Blu Hotel, Dakar Sea Plaza	Route de la Corniche O, Dakar 16868, Senegal	+221 33 869 33 33	-17.4730893,14.6945826,0
1	Radisson Hotel Dakar Diamniadio	Prolongement Autoroute a peage a cote du centr...	+221 32 824 48 48	-17.1970069,14.7385608,0
2	Paradialaw	Senegal, Toubab Dialao, toubab dialaw , prés d...	NaN	-17.14559,14.5963946,0
3	Hotel Club Royal Saly	Route Saly-Ngaparou M'bour SN, 23002, Senegal	+221 33 939 52 30	-17.0221955,14.4432474,0
4	Le Grand Hôtel de Thiés	Thies, Senegal	+221 33 951 00 98	-16.9280314,14.7950366,0
	hotel_names	addresses	phone_number	coordinates
count	30	30	28	30
unique	30	30	28	30
top	ibis Dakar	Avenue Nelson Mandela, Dakar, Senegal	+221 33 869 33 33	-17.0221955,14.4432474,0
freq	1	1	1	1
	pop_est	continent	name	iso_a3	gdp_md_est	geometry
0	920938	Oceania	Fiji	FJI	8374.0	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...
1	53950935	Africa	Tanzania	TZA	150600.0	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...
2	603253	Africa	W. Sahara	ESH	906.5	POLYGON ((-8.66559 27.65643, -8.66512 27.58948...
3	35623680	North America	Canada	CAN	1674000.0	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...
4	326625791	North America	United States of America	USA	18560000.0	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...