UNCERTML - DESCRIBING AND COMMUNICATING UNCERTAINTY Matthew Williams williamw@aston.ac.uk OVERVIEW Introduction. Motivation Webs. UncertML Use – the Semantic and Sensor overview. case – The INTAMAP project. Conclusions. MOTIVATION The semantic and sensor webs THE SENSOR WEB SENSOR WEB ENABLEMENT (SWE) Open Geospatial Consortium (OGC) initiative Interoperability interfaces and metadata encodings. <Quantity id="elevationAngle" fixed="false" Real time integration of heterogeneous sensor webs definition="urn:ogc:def:scanElevationAngle"> into the information infrastructure. <uom xlink:href="urn:ogc:unit:degree"/> <quality> <Tolerance definition="urn:ogc:def:tolerance2std"> Current SWE standards <value> -0.02 0.02 </value> </Tolerance> & Measurements Observations </quality> SensorML <value> 25.3 </value> </Quantity> SWE Common No formal standard for quantifying uncertainty HOW UNCERTAINTY IS USED WITHIN THE SEMANTIC WEB PR-OWL: a Bayesian Ontology Language for the Semantic Web: Extends OWL to allow probabilistic knowledge to be represented in an ontology. Used for reasoning with Bayesian inference. Random variables are described by either a PR-OWL table (discrete probability) or using a proprietary format. Other standards looking at similar concepts: BayesOWL. FuzzyOWL. What next? A formal open standard for quantifying complex uncertainties Extend to allow continuous distributions More powerful reasoning, richer representations UNCERTML OVERVIEW Split into three distinct packages (distributions, statistics & realisations). DISTRIBUTIONS <un:Distribution definition="http://dictionary.uncertml.org/distributions/gauss ian"> <un:parameters> <un:Parameter definition="http://dictionary.uncertml.org/distributions/gauss ian/mean"> <un:value>34.564</un:value> </un:Parameter> <un:Parameter definition="http://dictionary.uncertml.org/distributions/gauss ian/variance"> <un:value>67.45</un:value> </un:Parameter> </un:parameters> </un:Distribution> UNCERTML An overview WEAK VS. STRONG Weak-typed Benefits Strong-typed Benefits <Distribution Genericdefinition=“http://uncertml.org/gaussian”> features have Produces relatively <parameter definition=“http://uncertml.org/mean”>34.2</parameter> genericdefinition=“http://uncertml.org/variance”>12.4</parameter> properties – simple XML features <parameter </Distribution> extensible Drawbacks <GaussianDistribution> <mean>34.2</mean> Validation becomes <variance>12.4</variance> less meaningful </GaussianDistribution> Drawbacks Not easily extended – all domain features must be known a priori THE UNCERTML DICTIONARY Weak-typed designs rely on dictionaries. Includes definitions of key distributions & statistics. URIs link to dictionary entry and provide semantics. Could be written in Semantic Web standards (OWL, RDF etc). UNCERTML – DICTIONARY EXAMPLE <gml:Dictionary xmlns:gml="http://www.opengis.net/gml" gml:id="DISTRIBUTIONS"> <gml:name>All Probability Distributions</gml:name> <gml:description>Distributions dictionary</gml:description> <gml:dictionaryEntry> <un:DistributionDefinition xmlns:un="http://www.intamap.org/uncertml" gml:id="Gaussian"> <gml:description>Gaussian distribution</gml:description> <gml:name>Gaussian</gml:name> <gml:name>Normal</gml:name> <un:functions> <un:FunctionDefinition gml:id="Gaussian_Cumulative_Distribution_Function"> <gml:description>cumulative distribution function</gml:description> <gml:name>Cumulative Distribution Function</gml:name> <un:mathML> <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mfrac> <mml:mn>1</mml:mn> <mml:mn>2</mml:mn> </mml:mfrac> SEPARATION OF CONCERNS Several competing standards already exist addressing the issue of units and location. Geospatial information not always relevant – Systems biology. Do what we know – do it well! UNCERTML An applied case study THE INTAMAP PROJECT An automatic, interoperable service providing real time interpolation between observations. EURDEP providing radiological data as a case study. Provide real time predictions to aid risk management through a Web Processing Service interface. UNCERTML IN INTAMAP ‘Really clever’ Bayesian inference: Different sensor errors. Change of support. Fast & approximate algorithms. COMPARING PREDICTIONS WITH AND WITHOUT UNCERTML Without UncertML With UncertML CONCLUSIONS Currently no interoperable standard which fully describes random variables. UncertML provides an extensible, weak-typed, design that can quantify uncertainty using: Distributions. Statistics. Realisations. Provide richer information for use in decision support systems. <om:Observation> <un:DistributionArray> <om:procedure xlink:href="http://www.mydomain.com/sensor_models/temperature"/> <un:elementType> <om:resultQuality> <un:Distribution <un:Distribution definition="http://dictionary.uncertml.org/distributions/gaussian"> definition="http://dictionary.uncertml.org/distributions/gaussian"> <un:parameters> <un:parameters> <un:Parameter definition="http://dictionary.uncertml.org/distributions/gaussian/parameters/mean"> <un:Parameter <un:value>0.0</un:value> definition="http://dictionary.uncertml.org/distributions/gaussian/mean"/> </un:Parameter> <un:Parameter <un:Parameter definition="http://dictionary.uncertml.org/distributions/gaussian/variance"/> definition="http://dictionary.uncertml.org/distributions/gaussian/parameters/variance"> </un:parameters> <un:value>3.6</un:value> </un:Parameter> </un:Distribution> </un:parameters> </un:elementType> </un:Distribution> <un:elementCount>5</un:elementCount> </om:resultQuality> <swe:encoding> <om:observedProperty xlink:href="urn:x-ogc:def:phenomenon:OGC:AirTemperature"/> <swe:TextBlock decimalSeparator="." blockSeparator=" " <om:featureOfInterest> <sa:SamplingPoint> tokenSeparator=","/> <sa:sampledFeature xlink:href="http://www.mydomain.com/sampling_stations/ws-04231"/> </swe:encoding> <sa:position> <swe:values> <gml:Point> 35.2,56.75 <gml:pos srsName="urn:x-ogc:def:crs:EPSG:4326"> 31.2,65.31 52.4773635864 -1.89538836479 </gml:pos> 28.2,54.23 </gml:Point> 35.6,45.21 </sa:position> 41.5,85.24 </sa:SamplingPoint> </swe:values> </om:featureOfInterest> </un:DistributionArray> <om:result xsi:type="gml:MeasureType" uom="urn:ogc:def:uom:OGC:degC">19.4</om:result> </om:Observation> UNCERTML IN INTAMAP