The Federal Chancellery maintains data on the Swiss political system. The curia dataset is publicly available and it provides data on political parties, parliamentary comissions, members of parliament and their affiliations.
Parliament data is also available as Linked Data.
Swiss political data can be accessed with SPARQL queries.
You can send queries using HTTP requests. The API endpoint is https://lindas.admin.ch/query/.
import pandas as pd
import plotly.express as px
import plotly.graph_objects as go
from plotly.subplots import make_subplots
import datetime
import warnings
from graphly.api_client import SparqlClient
%matplotlib inline
pd.set_option('display.max_rows', 100)
warnings.filterwarnings('ignore')
# Uncomment to install dependencies in Colab environment
#!pip install git+https://github.com/zazuko/graphly.git
sparql = SparqlClient("https://int.lindas.admin.ch/query")
sparql.add_prefixes({
"schema": "<http://schema.org/>"
})
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.
# https://s.zazuko.com/5nEJGu
query = """
SELECT ?council ?name ?gender ?age ?start
FROM <https://lindas.admin.ch/fch/curia>
FROM <https://lindas.admin.ch/fch/rvov>
WHERE {
VALUES ?_council {<https://politics.ld.admin.ch/council/N> <https://politics.ld.admin.ch/council/S>}
?member a schema:Person ;
schema:name ?name;
schema:gender ?gender;
schema:birthDate ?birthday;
schema:memberOf ?role.
?role a schema:Role;
schema:member ?member;
schema:startDate ?start;
schema:memberOf ?_council.
?_council schema:name ?council.
FILTER (lang(?council) = "de")
FILTER NOT EXISTS { ?role schema:endDate ?end }
BIND( "2022-06-16"^^<http://www.w3.org/2001/XMLSchema#date> as ?today )
BIND(YEAR(?today)-YEAR(?birthday) as ?age)
}
ORDER BY ?age
"""
df = sparql.send_query(query)
df = df.replace({"http://schema.org/Male": "Male", "http://schema.org/Female": "Female"})
bucket=5
df["bucket"] = df.age.apply(lambda x: "{}-{}".format((x//bucket)*bucket, (x//bucket+1)*bucket))
counts = df.groupby(["bucket", "gender"]).size().reset_index(name='count')
fig = px.bar(counts, x="bucket", y="count", color="gender",
barmode="stack", color_discrete_sequence=[px.colors.qualitative.Plotly[1],px.colors.qualitative.Plotly[0]])
fig.update_layout(
title='Members of Parliament (as of today)',
title_x=0.5,
yaxis_title="Share in %",
xaxis_title="Age"
)
fig.show()
fig = make_subplots(rows=2, cols=1, shared_yaxes=True, y_title='Share in %', x_title='Age')
for i, gender in enumerate(df.gender.unique()[::-1]):
subset=df[df.gender == gender]["age"]
fig.append_trace(go.Histogram(x=subset, histnorm='percent', xbins=dict(start=25, end=80,size=5), name=gender), row=i+1, col=1)
fig.update_layout(
title='Members of Parliament (as of today)',
title_x=0.5,
bargap=0.1
)
fig.update_yaxes(range=[0,27.5])
fig.update_xaxes(range=[25,80])
fig.show()
counts = df.groupby(["council", "gender"]).size().reset_index(name="count")
fig = make_subplots(rows=1, cols=2, subplot_titles=counts["council"].unique(), specs=[[{"type": "pie"}, {"type": "pie"}]])
for i, council in enumerate(counts["council"].unique()):
fig.add_trace(go.Pie(
values=counts[counts.council == council]["count"],
labels=counts[counts.council == council]["gender"]
), row=1, col=i+1)
fig.update_annotations(yshift=-280)
fig.update_layout(height=400, title={"text": "Members of Parliament (as of today)", "x": 0.5})
fig.show()
# https://s.zazuko.com/3VYuVk
query = """
SELECT ?name ?gender ?birthday ?start ?end
FROM <https://lindas.admin.ch/fch/curia>
FROM <https://lindas.admin.ch/fch/rvov>
WHERE {
VALUES ?_council {<https://politics.ld.admin.ch/council/N> <https://politics.ld.admin.ch/council/S>}
?member a schema:Person ;
schema:name ?name;
schema:gender ?gender;
schema:birthDate ?birthday;
schema:memberOf ?role.
?role a schema:Role;
schema:member ?member;
schema:startDate ?start;
schema:memberOf ?_council.
OPTIONAL {?role schema:endDate ?end.}
}
ORDER BY ?start
"""
df = sparql.send_query(query)
df.end = df.end.fillna(datetime.datetime.now())
m = []
f = []
yrs = range(1965,2021)
for year in yrs:
point_in_time = datetime.datetime(year, 7, 1, 12, 0, 0)
subset = df[(df.start <= point_in_time) & (point_in_time < df.end)]
counts = subset.gender.value_counts()
m.append(counts["http://schema.org/Male"])
if "http://schema.org/Female" in counts:
f.append(counts["http://schema.org/Female"])
else:
f.append(0)
res = pd.DataFrame(data={"year": yrs, "male": m, "female": f})
res["f_share"] = res["female"]/(res["female"] + res["male"])
fig = make_subplots(rows=1, cols=1)
fig.append_trace(go.Bar(x=res["year"],y=(1-res["f_share"])*100,
name="male",
marker_color=px.colors.qualitative.Plotly[0]), row=1, col=1)
fig.append_trace(go.Bar(x=res["year"],y=(res["f_share"])*100,
name="female",
marker_color=px.colors.qualitative.Plotly[1]), row=1, col=1)
fig['layout']['yaxis']['title']='MPs share in %'
fig['layout']['yaxis']['range']= [0,100]
fig.update_layout(height=800, title={"text": "Female in Swiss Parliment", "x": 0.5}, barmode = "stack",
legend = {"x": 1, "y": 0.37})
fig.show()
yrs = range(1850, 2021, 4)
#bucket=10
#all_buckets = set(range(2,9))
bucket=5
all_buckets = set(range(5,17))
bucket2age = {i: "{}-{}".format(i*bucket, (i+1)*bucket-1) for i in all_buckets}
res = pd.DataFrame()
average_age = pd.DataFrame(columns=["year", "Male", "Female"])
for year in yrs:
point_in_time = datetime.datetime(year, 7, 1, 12, 0, 0)
subset = df[(df.start <= point_in_time) & (point_in_time < df.end)]
subset.loc[:,"age"] = point_in_time.year - subset.birthday.dt.year
males_only = subset['gender'] == 'http://schema.org/Male'
average_age_male = subset.loc[males_only, 'age'].mean()
females_only = subset['gender'] == 'http://schema.org/Female'
average_age_female = subset.loc[females_only, 'age'].mean()
#average_age = subset["age"].mean()
subset.loc[:,"bucket"] = subset.age.apply(lambda x: x//bucket)
grouped = subset.groupby(["bucket"]).size().reset_index(name='count')
grouped.loc[:, "count"] = grouped["count"]/(grouped["count"].sum())*100
for b in all_buckets.difference(set(grouped.bucket.unique())):
grouped = grouped.append({"bucket": b, "count": 0.0}, ignore_index=True)
grouped.loc[:, "year"] = year
res = res.append(grouped)
average_age.loc[len(average_age.index)] = [year, average_age_male, average_age_female]
res = res.sort_values(by=["year", "bucket"]).reset_index(drop=True)
res.loc[:,"age"] = res["bucket"].replace(bucket2age)
fig = px.bar(res, x="age", y="count", animation_frame="year", range_y=[0,35])
fig.update_layout(
title='Age Distribution',
title_x=0.5,
yaxis_title="Share in %",
xaxis_title="Age",
)
fig.show()
fig_average = px.bar(average_age, x="year", y=["Male", "Female"], barmode="group")
fig_average.update_layout(
title='Age over time',
title_x=0.5,
yaxis_title="Average Age",
xaxis_title="Year",
legend=dict(title=None)
)
fig_average.show()