import
 numpy
 as
 np

import
 pandas
 as
 pd

import
 sparql


service
 =
 sparql
.
Service
(
"https://query.wikidata.org/sparql"
,
 method
=
"GET"
)

statement
 =
 """

PREFIX wikibase: <http://wikiba.se/ontology#>

PREFIX wd: <http://www.wikidata.org/entity/> 

PREFIX wdt: <http://www.wikidata.org/prop/direct/>

PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>

PREFIX p: <http://www.wikidata.org/prop/>

PREFIX v: <http://www.wikidata.org/prop/statement/>


SELECT ?artwork ?artworkLabel ?motiv ?motivLabel WHERE {	

   ?artwork wdt:P195 wd:Q671384 .        

   ?artwork wdt:P180 ?motiv . 

   SERVICE wikibase:label {

     bd:serviceParam wikibase:language "da,en" .

   } 

 } 

"""


result
 =
 service
.
query
(
statement
)

df
 =
 pd
.
DataFrame
(
result
.
fetchall
(),
 columns
=
result
.
variables
)


# Set up feature matrix

motivs
 =
 set
([
item
.
value
 for
 item
 in
 df
[
'motivLabel'
]])

artworks
 =
 set
([
item
.
value
 for
 item
 in
 df
[
'artwork'
]])

feature_matrix
 =
 pd
.
DataFrame
(
0
,
 index
=
artworks
,
 columns
=
motivs
)

artwork_mapper
 =
 {}

for
 n
,
 row
 in
 df
.
iterrows
():

    artwork
 =
 row
[
'artwork'
]
.
value

    motiv
 =
 row
[
'motivLabel'
]
.
value

    feature_matrix
.
ix
[
artwork
,
 motiv
]
 =
 1
 
    artwork_mapper
[
artwork
]
 =
 row
[
'artworkLabel'
]
.
value


# Possible scaling

feature_matrix
 =
 feature_matrix
.
divide
(
np
.
sqrt
(
feature_matrix
.
sum
(
axis
=
1
)),
 axis
=
'index'
)


# Machine learning decomposition

from
 sklearn.decomposition
 import
 NMF

decomposer
 =
 NMF
(
n_components
=
2
)

decomposer
.
fit
(
feature_matrix
)

transformed
 =
 decomposer
.
transform
(
feature_matrix
)

>>>
 feature_matrix
.
columns
[(
-
decomposer
.
components_
[
0
,
 :])
.
argsort
()[:
10
]]

Index
([
u
'hund'
,
 u
'mand'
,
 u
'kvinde'
,
 u
'træ'
,
 u
'bord'
,
 u
'kurv'
,
 u
'borddug'
,
 u
'barn'
,
 u
'hest'
,
 u
'Sværd, sværd'
],
 dtype
=
'object'
)

>>>
 feature_matrix
.
columns
[(
-
decomposer
.
components_
[
1
,
 :])
.
argsort
()[:
10
]]

Index
([
u
'himmel'
,
 u
'sky'
,
 u
'mand'
,
 u
'vej'
,
 u
'Hede, hede'
,
 u
'Lyng, lyng'
,
 u
'træ'
,
 u
'Bjerg, bjerg'
,
 u
'Vandfald, vandfald'
,
 u
'Far, far'
],
 dtype
=
'object'
)


>>>
 [
artwork_mapper
[
q
]
 for
 q
 in
 feature_matrix
.
index
[(
-
transformed
[:,
 0
])
.
argsort
()[:
10
]]]

[
u
'Dansebakken ved Sorgenfri'
,
 u
'Johannes D
\xf8
berens henrettelse'
,
 u
'B
\xf8
geskov i maj. Motiv fra Iselingen'
,
 u
'Frederik II bygger Kronborg'
,
 u
'Frederik II bygger Kronborg'
,
 u
'En hyrdinde med en hund'
,
 u
'Fra et romersk osteria'
,
 u
'Et selskab af danske kunstnere i Rom'
,
 u
'Susanne og de to 
\xe6
ldste'
,
 u
'Bella and Hanna. The Eldest Daughters of M. L. Nathanson'
]

>>>
 [
artwork_mapper
[
q
]
 for
 q
 in
 feature_matrix
.
index
[(
-
transformed
[:,
 1
])
.
argsort
()[:
10
]]]

[
u
'En jysk f
\xe5
rehyrde p
\xe5
 heden'
,
 u
'Landskab med vandfald og hytte'
,
 u
'Maleren i landsbyen'
,
 u
'Udsigt gennem tre af de nordvestlige buer i Colosseums tredje stokv
\xe6
rk'
,
 u
'Udsigt fra Kronborg Vold over flagbatteriet og Sundet til den svenske kyst'
,
 u
'Bjerglandskab med en flod'
,
 u
'Hedestudie'
,
 u
'Landscape with Two Horses'
,
 u
'Udsigt fra Parthenon p
\xe5
 Akropolis'
,
 u
'Birketr
\xe6
er p
\xe5
 L
\xe6
s
\xf8
 efter regn'
]