This HTML5 document contains 5 embedded RDF statements represented using HTML+Microdata notation.

The embedded RDF content will be recognized by any processor of HTML5 Microdata.

Namespace Prefixes

PrefixIRI
n4http://data.silknow.org/actor/XGBoost-classifier/
ecrmhttp://erlangen-crm.org/current/
rdfhttp://www.w3.org/1999/02/22-rdf-syntax-ns#
provhttp://www.w3.org/ns/prov#
n6http://data.silknow.org/object/
xsdhhttp://www.w3.org/2001/XMLSchema#
n2http://data.silknow.org/activity/

Statements

Subject Item
n2:ef24c7ee-ff23-51f4-aea8-1c4a3fcd3486
rdf:type
prov:Activity
prov:used
n6:f3913dfd-74c9-31d6-9d7d-da1ded9c7357
ecrm:P70_documents
Predictions made using a machine learning algorithm called Gradient Tree Boosting. The software predicts values for four properties, namely 'production timespan', 'production place', 'technique' and 'material'. The predictions are based on the values that are already present for these properties together with the museum name. It is therefore used to fill 'gaps', i.e. missing values for some properties, in the records of the Knowledge Graph and trained with data already present in it. This version of the models is trained on a snapshot of the Knowledge Graph from May 2021. The single task multi-class classification (mutually exclusive classes) models make use of a softmax function, the maximum value output of which corresponds to the provided prediction score.
prov:atTime
2021-05-21T00:00:00
prov:wasAssociatedWith
n4: