Post on 01-Apr-2015
Using Ontologies in the Development of an Innovating System for Elderly
People Tele-Assistance
URL: http://siul02.si.ehu.es/
Outline
Goal System architecture Domain Ontology. MedOnt Operational ontology. OperOnt Conclusions Future Work
Goal of the system
To provide high quality assistance to elderly people, giving them more freedom and protection.
This system uses agent technology, semantic web technology and personal Digital Assistant (PDA) with wireless communications.
Our Proposal AINGERU:
– offers an active assistance by using agents that behave in face of anomalous situations;
– offers an anywhere and anytime assistance by using wireless communications and PDAs; and,
– allows to monitor vital signs by using sensors that capture the values of those signus and feed a decision support system that analyzes them and generates an alarm when necessary.
System Architecture
Main elements of the system Two logic-based Ontologies
MedOnt : Domain Ontology
OperOnt: Operational Ontology
Software AgentsThey help to manage the distributed and heterogeneous
framework
They can accomplish tasks autonomously and can interact with other agents
Web ServicesAllow relatives and physicians related to a monitored
person to consult data about user appointments, medicines taken etc.
MedOnt. Goal
to describe the different situations in which a medical alarm has to be activated
In this ontology are described the different symptoms that a user can have, with respect to Vital Constants that several sensors can monitor, and also the usual illnesses that elderly people suffer from.
MedOnt. Main features
• This ontology can be customized for every user. That is, this ontology can be easily adapted to different users depending on their own situation.
• This ontology is being developed by experts (so far we only have a small prototype).
• Described using OWL
OperOnt. Goal We wanted a specification of the system that
would facilitate:1. The interoperability of AINGERU2. The understanding of AINGERU features by external
actors3. The AINGERU evolution
String XML (DTD XML Schema) DAML+OIL (OWL) Ontología
OperOnt: To describe the operational model of AINGERU
OperOnt. Main features The concepts in this ontology are defined independently
from any agent system implementation, so that it helps interoperability among agents without pre-defined agreements.
At the same time, the OperOnt ontology describes contextual information that several agents are able to share (understanding).
This ontology is easily extensible as the functionality of AINGERU increases (evolution)
OperOnt. Steps to arrive to it
Branches:– Actor (Human agents, Software agents, Web
services)– Message (according to their functionality in
FIPA)– Subject (terms about the topic in which the
message is centered)
OperOnt. Decription
OperOnt. Description
OperOnt. Decription
Wait
Assert(OperOnt,m1)
Message m1
<OperOnt:Query>
<OperOnt:ident>m1</OperOnt:ident>
<OperOnt:theSender><OperOnt:ConditionCheckerAgent /></OperOnt:theSender>
<OperOnt:theReceiver><OperOnt:LocationAgent /></OperOnt:theReceiver>
<OperOnt:tryNumber>1</OperOnt:tryNumber>
<OperOnt:theSubject><OperOnt:Location /></OperOnt:theSubject>
<OperOnt:priority>1</OperOnt:priority>
</OperOnt:Query>
M1 LocationQuery
<OperOnt:Query>
<OperOnt:ident>m1</OperOnt:ident>
<OperOnt:theSender><OperOnt:ConditionCheckerAgent /></OperOnt:theSender>
<OperOnt:theReceiver><OperOnt:LocationAgent /></OperOnt:theReceiver>
<OperOnt:tryNumber>1</OperOnt:tryNumber>
<OperOnt:theSubject><OperOnt:Location /></OperOnt:theSubject>
<OperOnt:priority>1</OperOnt:priority>
</OperOnt:Query>
<owl:Class>
<owl:intersectionOf parseType="owl:collection">
<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/messages#Query"/>
<owl:Restriction>
<owl:onProperty rdf:resource="file:/C:/Aingeru/OperOnt/messages#theSubject"/>
<owl:someValuesFrom>
<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/subjects#Location"/>
</owl:someValuesFrom>
</owl:Restriction>
</owl:intersectionOf>
<rdfs:subClassOf>
<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/messages#LocationQuery"/>
</rdfs:subClassOf>
</owl:Class>
Wait
Assert(OperOnt,m1)
L=GetLocation()Assert(OperOnt, Inform, subject.Location)
Message m1
LocationQuery
M2 Inform. M2.theSubject Location<owl:Class>
<owl:intersectionOf parseType="owl:collection">
<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/messages#Inform"/>
<owl:Restriction>
<owl:onProperty rdf:resource="file:/C:/Aingeru/OperOnt/messages#theSubject"/>
<owl:someValuesFrom>
<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/subjects#Location"/>
</owl:someValuesFrom>
</owl:Restriction>
</owl:intersectionOf>
<rdfs:subClassOf>
<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/messages#LocationInform"/>
</rdfs:subClassOf>
</owl:Class>
M2 LocationInform
M2 LocationInform
<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/messages#LocationInform">
<rdfs:label>LocationInform</rdfs:label>
<rdfs:subClassOf>
<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/messages#Inform-result/>
</rdfs:subClassOf>
</owl:Class>
<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/messages#Inform-result">
<rdfs:label>Inform-result</rdfs:label>
<rdfs:subClassOf>
<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/messages#Inform"/>
</rdfs:subClassOf>
<rdfs:subClassOf>
<owl:Restriction>
<owl:onProperty rdf:resource="file:/C:/Aingeru/OperOnt/messages#theInform"/>
<owl:someValuesFrom rdf:resource="http://www.w3.org/2000/10/XMLSchema#string"/>
</owl:Restriction>
</rdfs:subClassOf>
</owl:Class>
M2 LocationInform Inform-result
M2 LocationInform Inform-result Inform<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/messages#Inform">
<rdfs:label>Inform</rdfs:label>
<rdfs:subClassOf>
<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/messages#Message"/>
</rdfs:subClassOf>
<rdfs:subClassOf>
<owl:Restriction>
<owl:onProperty rdf:resource="file:/C:/Aingeru/OperOnt/messages#theSubject"/>
<owl:someValuesFrom>
<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/subjects#Subject"/>
</owl:someValuesFrom>
</owl:Restriction>
</rdfs:subClassOf>
<rdfs:subClassOf>
<owl:Restriction>
<owl:onProperty rdf:resource="file:/C:/Aingeru/OperOnt/messages#toWhich"/>
<owl:someValuesFrom>
<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/subjects#Message"/>
</owl:someValuesFrom>
</owl:Restriction>
</rdfs:subClassOf>
</owl:Class>
M2 LocationInform Inform-result Inform Message
<owl:Class rdf:about=
"file:/C:/Aingeru/OperOnt/messages#Message">
<rdfs:label>Message</rdfs:label>
<rdfs:subClassOf>
<owl:Restriction>
<owl:onProperty rdf:resource=
"file:/C:/Aingeru/OperOnt/messages#tryNumber"/>
<owl:someValuesFrom rdf:resource=
"http://www.w3.org/2000/10/XMLSchema#integer"/>
</owl:Restriction>
</rdfs:subClassOf>
<rdfs:subClassOf>
<owl:Restriction>
<owl:onProperty rdf:resource=
"file:/C:/Aingeru/OperOnt/messages#ident"/>
<owl:someValuesFrom rdf:resource=
"http://www.w3.org/2000/10/XMLSchema#string"/>
</owl:Restriction>
</rdfs:subClassOf>
<rdfs:subClassOf>
<owl:Restriction>
<owl:onProperty rdf:resource="file:/C:/Aingeru/OperOnt/messages#theSender"/>
<owl:someValuesFrom>
<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/actors#Actor"/>
</owl:someValuesFrom>
</owl:Restriction>
</rdfs:subClassOf>
<rdfs:subClassOf>
<owl:Restriction>
<owl:onProperty rdf:resource="file:/C:/Aingeru/OperOnt/messages#theReceiver"/>
<owl:someValuesFrom>
<owl:Class rdf:about="file:/C:/Aingeru/OperOnt/actors#Actor"/>
</owl:someValuesFrom>
</owl:Restriction>
</rdfs:subClassOf>
</owl:Class>
Roles: – tryNumber– ident– theSender– theReceiver– theSubject – theInform– toWhich
M2 LocationInform Inform-result Inform Message
Wait
Assert(OperOnt,m1)
L=GetLocation()Assert(OperOnt, Inform, subject.Location)
Send(m2)
Message m1
LocationQuery
LocationInform<OperOnt:Inform>
<OperOnt:ident>m2</OperOnt:ident>
<OperOnt:theSender><OperOnt: LocationAgent /></OperOnt:theSender>
<OperOnt:theReceiver>
<OperOnt: ConditionCheckerAgent />
</OperOnt:theReceiver>
<OperOnt:tryNumber>1</OperOnt:tryNumber>
<OperOnt:theSubject><OperOnt:Location /></OperOnt:theSubject>
<OperOnt:toWhich>m1</OperOnt:toWhich>
<OperOnt:theInform>43º18’26’’, -2º0’41’’</OperOnt:theInform>
</OperOnt:Inform>
Conclusions AINGERU, gives one step forward in the direction
of helping elderly people to increase their quality of life, by:– allowing the monitoring of people anywhere and
anytime. – Moreover, this monitoring is reactive. In order to
perform the detection process, the system uses an application domain ontology, MedOnt.
– Another contribution of the system is the development and management of an operational ontology, OperOnt, that permits agents that take part in the system to communicate at a semantic level.
Future Work
At the system level– To add new functionalities to the system. To
develop new agents.– To improve the users locatization system
At the ontologies level– To add new terms at both ontologies– To run a –DAM+OIL, OWL -reasoner system
at the PDAs