Wikidata Samian Ware Sites Visualisation

This notebook fetches Samian Ware data using a SPARQL query from Wikidata and creates bar charts.

Define SPARQL query service

from SPARQLWrapper import SPARQLWrapper, JSON
import pandas as pd
import matplotlib.pyplot as plt

def querySparql(query):
    sparql = SPARQLWrapper("https://query.wikidata.org/sparql")
    sparql.setQuery(query)
    sparql.setReturnFormat(JSON)
    results = sparql.queryAndConvert()
    return results['results']['bindings']

Define the SPARQL Query

# SPARQL Query for Samian Ware Kiln Sites
samianQuery = """
SELECT ?item ?samian ?itemLabel ?geo ?layerLabel ?layer ?kilnregion ?kilnregionLabel WHERE {
  ?item wdt:P31 wd:Q102202026;
    wdt:P361 wd:Q90412636;
    wdt:P706 ?kilnregion;
    wdt:P2888 ?samian;
    wdt:P625 ?geo;
    wdt:P706 ?layer.
  SERVICE wikibase:label { bd:serviceParam wikibase:language "en". }
}
"""

Fetch Data and Convert to DataFrame

# Fetch data using the SPARQL query
sparql_results = querySparql(samianQuery)

# Convert SPARQL JSON results into a DataFrame
data = []
for result in sparql_results:
    geo = result.get('geo', {}).get('value', None)
    lat, lon = None, None
    if geo:
        coords = geo.replace("Point(", "").replace(")", "").split()
        if len(coords) == 2:
            lon, lat = map(float, coords)
    data.append({
        "item": result.get("item", {}).get("value", None),
        "itemLabel": result.get("itemLabel", {}).get("value", None),
        "layerLabel": result.get("layerLabel", {}).get("value", None),
        "kilnregionLabel": result.get("kilnregionLabel", {}).get("value", None),
        "latitude": lat,
        "longitude": lon,
    })

df = pd.DataFrame(data)
df

	item	itemLabel	layerLabel	kilnregionLabel	latitude	longitude
0	http://www.wikidata.org/entity/Q103136249	Sagone (Samian Ware Productioncentre)	Italian (Samian Ware Kilnregion)	Italian (Samian Ware Kilnregion)	42.112142	8.702547
1	http://www.wikidata.org/entity/Q103136253	Albi (Samian Ware Productioncentre)	South Gaulish (Samian Ware Kilnregion)	South Gaulish (Samian Ware Kilnregion)	43.927186	2.149255
2	http://www.wikidata.org/entity/Q103136256	Rodez (Samian Ware Productioncentre)	South Gaulish (Samian Ware Kilnregion)	South Gaulish (Samian Ware Kilnregion)	44.352370	2.572616
3	http://www.wikidata.org/entity/Q102763431	La Graufesenque (Samian Ware Productioncentre)	South Gaulish (Samian Ware Kilnregion)	South Gaulish (Samian Ware Kilnregion)	44.100000	3.083333
4	http://www.wikidata.org/entity/Q103135455	Banassac (Samian Ware Productioncentre)	South Gaulish (Samian Ware Kilnregion)	South Gaulish (Samian Ware Kilnregion)	44.433333	3.200000
...	...	...	...	...	...	...
98	http://www.wikidata.org/entity/Q103136223	Torrita di Siena (Samian Ware Productioncentre)	Italian (Samian Ware Kilnregion)	Italian (Samian Ware Kilnregion)	43.166668	11.766666
99	http://www.wikidata.org/entity/Q103136227	Vasanello (Samian Ware Productioncentre)	Italian (Samian Ware Kilnregion)	Italian (Samian Ware Kilnregion)	42.416699	12.350000
100	http://www.wikidata.org/entity/Q103136235	Venosa (Samian Ware Productioncentre)	Italian (Samian Ware Kilnregion)	Italian (Samian Ware Kilnregion)	40.966667	15.816667
101	http://www.wikidata.org/entity/Q103136239	Micasasa (Samian Ware Productioncentre)	Dacian (Samian Ware Kilnregion)	Dacian (Samian Ware Kilnregion)	46.086660	24.108056
102	http://www.wikidata.org/entity/Q103136244	Lyon (Samian Ware Productioncentre)	Italian (Samian Ware Kilnregion)	Italian (Samian Ware Kilnregion)	45.760000	4.840000

103 rows × 6 columns

Visualise the Data with Charts

# Count sites by region and layer for stacked bar chart
region_layer_counts = df.groupby(["kilnregionLabel", "layerLabel"]).size().unstack(fill_value=0)

# Create pie chart data
region_counts = df["kilnregionLabel"].value_counts()

# Stacked Bar Chart
plt.figure(figsize=(12, 8))
region_layer_counts.plot(kind="bar", stacked=True, figsize=(12, 8), colormap="tab20")
plt.title("Distribution of Kiln Sites by Region and Layer", fontsize=16)
plt.xlabel("Kiln Region", fontsize=14)
plt.ylabel("Number of Sites", fontsize=14)
plt.xticks(rotation=45, ha="right")
plt.legend(title="Layer", bbox_to_anchor=(1.05, 1), loc="upper left")
plt.tight_layout()
plt.show()

<Figure size 1200x800 with 0 Axes>

# Pie Chart
plt.figure(figsize=(10, 10))
region_counts.plot(kind="pie", autopct="%1.1f%%", startangle=90, colors=plt.cm.Paired.colors)
plt.title("Proportion of Kiln Sites by Region", fontsize=16)
plt.ylabel("")  # Remove default ylabel
plt.tight_layout()
plt.show()