Swiss commerce register¶
The Federal Office of Justice maintains ZEFIX, the swiss commerce register for all legally operating businesses.
The register provides us with company name, type, description, and address.
ZEFIX is also available as Linked Data.
Setup¶
SPARQL endpoints¶
For companies data¶
Swiss commerce register data can be accessed with SPARQL queries.
You can send queries using HTTP requests. The API endpoint is https://lindas.admin.ch/query/.
For geodata¶
To understand companies' location, we will work with Swiss geodata. It can be accessed with GeoSPARQL under https://ld.geo.admin.ch/query.
import json
import os.path
import folium
import mapclassify
import matplotlib as mpl
import matplotlib.pyplot as plt
import pandas as pd
import geopandas as gpd
import plotly.express as px
import plotly.graph_objects as go
from plotly.subplots import make_subplots
from graphly.api_client import SparqlClient
%matplotlib inline
# Uncomment to install dependencies in Colab environment
#!pip install mapclassify
#!pip install git+https://github.com/zazuko/graphly.git
sparql = SparqlClient("https://lindas.admin.ch/query")
geosparql = SparqlClient("https://ld.geo.admin.ch/query")
sparql.add_prefixes({
"schema": "<http://schema.org/>",
"admin": "<https://schema.ld.admin.ch/>"
})
geosparql.add_prefixes({
"dct": "<http://purl.org/dc/terms/>",
"geonames": "<http://www.geonames.org/ontology#>",
"schema": "<http://schema.org/>",
"geosparql": "<http://www.opengis.net/ont/geosparql#>",
})
SPARQL queries can become very long. To improve the readibility, we will work wih prefixes.
Using the add_prefixes method, we can define persistent prefixes.
Every time you send a query, graphly will now automatically add the prefixes for you.
Companies by company type¶
Swiss law offers many opportnities to business owners. They can choose from several legal entites. What entities can we find in ZEFIX? Which ones are the most popular among enterpreneurs?
query = """
SELECT ?type (COUNT(DISTINCT ?company_iri) AS ?ccount)
WHERE {
?company_iri a admin:ZefixOrganisation.
?company_iri schema:additionalType/schema:name ?type .
FILTER(LANG(?type) = "de")
}
GROUP BY ?type
ORDER BY DESC(?ccount)
"""
df = sparql.send_query(query)
# Let's rename variables to english
de2en = {'Kommanditgesellschaft': "Limited Partnership",
'Ausländische Niederlassung im Handelsregister eingetragen': "Foreign Branch",
'Genossenschaft': "Cooperative",
'Verein': "Association",
'Kollektivgesellschaft': "General Partnership",
'Schweizerische Zweigniederlassung im Handelsregister eingetragen': "Swiss Branch",
'Stiftung': "Foundation",
'Einzelunternehmen': "Sole proprietorship",
'Aktiengesellschaft': "Joint-stock Company",
'Gesellschaft mit beschränkter Haftung GMBH / SARL': "Limited Liability Company"}
df["type"] = df["type"].apply(lambda x: de2en[x] if x in de2en else x)
fig = px.bar(df[df.ccount > 500], y="type", x="ccount", orientation = "h", labels={"type": "", "ccount": "Company Count"})
fig.update_layout(
title='Which company type is most popular?',
title_x=0.5,
)
fig.update_layout(bargap=0.40)
fig.show()
Companies by municipality¶
The economic activity varies significantly across regions. What are the most popular places to register a company?
query = """
SELECT ?municipality ?muni_id (COUNT(?sub) AS ?companies)
FROM <https://lindas.admin.ch/foj/zefix>
FROM <https://lindas.admin.ch/territorial>
WHERE {
?sub a admin:ZefixOrganisation ;
admin:municipality ?muni_id.
?muni_id schema:name ?municipality;
}
GROUP BY ?municipality ?muni_id
ORDER BY DESC(?companies)
"""
df = sparql.send_query(query)
# Communes
query = """
SELECT ?muni_id ?population ?boundary
WHERE {
?muni_iri dct:hasVersion ?version ;
geonames:featureCode geonames:A.ADM3 .
?version schema:validUntil "2020-12-31"^^<http://www.w3.org/2001/XMLSchema#date>;
geonames:population ?population;
geosparql:hasGeometry/geosparql:asWKT ?boundary.
BIND(IRI(REPLACE(STR(?muni_iri), "https://ld.geo.admin.ch/boundaries/", "https://ld.admin.ch/")) AS ?muni_id)
}
"""
communes = geosparql.send_query(query)
communes = communes.set_crs(epsg=4326)
communes["center"] = communes.to_crs(epsg=3035).centroid.to_crs(epsg=4326)
join = pd.merge(communes, df, how="inner", on="muni_id")
join.sort_values(by="companies", ascending=False, inplace=True)
join = join.reset_index(drop=True)
join[["municipality", "companies"]]
The most popular places are the biggest cities.
This should come at no surprise. The more people live in a certain area, the more business founders it will have. The company density is expected to be directly correlated with population density.
Now, let's take into account the population density. Which regions have most companies per 10.000 inhabitants?
join["companies_per_inhabitants"] = round(join.companies/join.population*10000)
join["companies_per_inhabitants"] = join["companies_per_inhabitants"].astype(int)
join.sort_values(by="companies_per_inhabitants", ascending=False, inplace=True)
join = join.reset_index(drop=True)
join[["municipality", "companies_per_inhabitants"]].head()
Interesting! Zug is the only bigger city that appears on the top of the list.
Where are the most popular registration destinations?
# Companies per 10.000 inhabitants
style_function = lambda x: {'fillColor': '#ffffff',
'color':'#000000',
'fillOpacity': 0.1,
'weight': 0.1}
highlight_function = lambda x: {'fillColor': '#989898',
'color':'#000000',
'fillOpacity': 0.8}
def plot_heatmap(df, viz_column, hover_cols, col2label, title):
classifier = mapclassify.NaturalBreaks(y=df[viz_column], k=5)
bins = [df[viz_column].min()] + list(classifier.bins)
cols = ["municipality", viz_column, *hover_cols]
m = folium.Map(location=[46.83, 8.13], zoom_start=8, tiles="cartodbpositron")
folium.Choropleth(
geo_data=json.loads(df.to_json()),
data=df,
columns=cols,
key_on="feature.properties.municipality",
fill_color="YlOrRd",
fill_opacity=1,
line_weight=0,
smooth_factor=0,
bins=bins,
reset=True,
legend_name=col2label[viz_column],
).add_to(m)
hover = folium.features.GeoJson(
df,
style_function=style_function,
control=False,
highlight_function=highlight_function,
tooltip=folium.features.GeoJsonTooltip(
fields=cols,
aliases=[col2label[i] for i in cols],
style=("background-color: white; color: #333333; font-family: arial; font-size: 12px; padding: 10px;")
)
)
folium.LayerControl().add_to(m)
m.add_child(hover)
m.keep_in_front(hover)
title_html = '''<h3 align="center" style="font-size:16px"><b>{}</b></h3>'''.format(title)
m.get_root().html.add_child(folium.Element(title_html))
return m
join = join.drop(columns="center")
col2label = {"municipality": "Municipality", "companies_per_inhabitants": "Companies per 10.000 inhabitants", "companies": "Companies ", "population": "Population"}
plot_heatmap(join, "companies_per_inhabitants", ["companies", "population"], col2label, "Where are most companies registered?")