Smart Applications REFerence ontology (SAREF)

SAREF is a comprehensive suite of interrelated ontologies that defines a shared model of consensus for enabling semantic interoperability across IoT solutions and smart applications from diverse providers and industry sectors. It provides standardized vocabulary for describing smart devices, their capabilities, interactions, and services in Internet of Things (IoT) and smart environments (smart homes, smart buildings, smart cities). SAREF captures essential IoT concepts including devices, sensors, actuators, services, properties, and the relationships between them, facilitating machine-to-machine communication and automation. The SAREF suite is published as open standards by ETSI Technical Committee SmartM2M (Smart Machine-to-Machine communications), ensuring broad industry adoption and compatibility. SAREF enables semantic data spaces by providing common terminology that allows data and services from different IoT platforms and manufacturers to be integrated and understood.

Example Usage: Annotate an IoT device deployment with SAREF terms describing the device type (smart thermostat, light bulb, motion sensor), its capabilities (temperature measurement, brightness control), services offered (remote adjustment, scheduling), and relationships to control systems and user preferences.

Metrics & Statistics

Graph Statistics

Total Nodes

804

Total Edges

1720

Root Nodes

14

Leaf Nodes

376
Knowledge Coverage Statistics

Classes

129

Individuals

10

Properties

89
Hierarchical Metrics

Maximum Depth

1

Minimum Depth

0

Average Depth

0.07

Depth Variance

0.06
Breadth Metrics

Maximum Breadth

14

Minimum Breadth

1

Average Breadth

7.50

Breadth Variance

42.25
LLMs4OL Dataset Statistics

Term Types

10

Taxonomic Relations

88

Non-taxonomic Relations

0

Average Terms per Type

10.00

Usage Example

Use the following code to import this ontology programmatically:

from ontolearner.ontology import SAREF

ontology = SAREF()
ontology.load("path/to/SAREF-ontology.rdf")

# Extract datasets
data = ontology.extract()

# Access specific relations
term_types = data.term_typings
taxonomic_relations = data.type_taxonomies
non_taxonomic_relations = data.type_non_taxonomic_relations