Skip to content

KGHeartBeat API is a Python library which allows developers to perform quality analysis of knowledge graphs

Notifications You must be signed in to change notification settings

GabrieleT0/KGHeartbeat-API

Repository files navigation

KGHeartBeat API

  1. What is KGHeartBeat API
  2. Installation
  3. Examples

What is KGHeartBeat API

Library that allows you to perform Knowledge Graph (Linked Open Data) quality analysis.

The granularity with which KG quality can be measured is at the metric level, meaning that it is possible to combine the calculation of multiple metrics to obtain KG quality for a dimension. The examples section shows how you can calculate the quality of different dimensions combining together the calculation of different metrics.

For documentation on the library implementation visit here: https://gabrielet0.github.io/KGHeartbeat-API/reference/

For information on which metrics are supported and how they are calculated visit here: https://isislab-unisa.github.io/KGHeartbeat/

Installation

pip install kgheartbeat

Examples

  1. Calculate the Availability dimension
  2. Calculate the Licensing dimension
  3. Calcuate the Versatility dimension

Calculate the Availability dimension

from kgheartbeat import KnowledgeGraph

# Instanziate a KnowledgeGraph class, passing the id of the kg to be analyzed
kg = KnowledgeGraph('dbpedia')

# Check the SPARQL endpoint availability
sparqlAv = kg.checkEndpointAv()
# Check if the links for download the dataset is up
checkDump = kg.checkDownload()
# Check if there are any inactive links
inactiveLks = kg.checkInactiveLinks()
# Calculate the URI's deferenceability (based on 5000 randomly recoverable uri). THIS COULD TAKE TIME, DEPENDS ON THE SPEED OF THE ENDPOINT (~45 min. for DBpedia)
uriDef = kg.getURIsDef()

#Print all the results
print(f"SPARQL endpoint availability: {sparqlAv}\n \
       RDF dump link availability: {checkDump}\n\
       Any inactive links: {inactiveLks}\n\
       URIs deferenceability: {uriDef}")

Calculate the Licensing dimension

from kgheartbeat import KnowledgeGraph

# Instanziate a KnowledgeGraph class, passing the id of the kg to be analyzed
kg = KnowledgeGraph('taxref-ld')

#Search for the machine-redeable license
mr_license = kg.getLicenseMR()
#Search for a human-redeable license
hr_license = kg.getLicenseHR()

print(f"Machine redeable license: {mr_license}\nHuman-redeable license: {hr_license}")

Calcuate the Versatility dimension

# Instanziate a KnowledgeGraph class, passing the id of the kg to be analyzed
kg = KnowledgeGraph('bncf-ns')

#Find the different serialization formats available (e.g. .rdf, .n3, .xml)
formats = kg.getSerializationFormat()
#Get languages if is indicated
languages = kg.getLanguages()
#get all the links to access the KG
link_access = kg.getAccessAtKG()

print(f"Serialization formats: {formats}\nLanguages: {languages}\n Link to access the KG:{link_access}\n")

About

KGHeartBeat API is a Python library which allows developers to perform quality analysis of knowledge graphs

Resources

Stars

Watchers

Forks

Languages