Exmaple: data quality management of tweets.

XML file over REST API (HTTP POST) for creation of quality policies. An organizational quality policy, 
a filtering policy, and a search filtering policy are created:

<PolicyCollection>
  <OrganizationalQualityPolicy>
    <Identifier>10</Identifier>
    <DataSourceType>Twitter</DataSourceType> # must match with DataSourceType of metadata 
    <MetadataId>metadata1</MetadataId> # must match with Identifier of metadata 
    <QualityAttributes>1,2,3</QualityAttributes> 
   <FilteringPolicy>3</FilteringPolicy> # must match filtering policy id
  </OrganizationalQualityPolicy>
  <FilteringPolicy>
    <Identifier>3</Identifier>
    <Description>Filter for sentiment analysis</Description>
    <QualityAttributeRanges>1:0.5-1.0,2:0.5-1.0</QualityAttributeRanges> 
  </FilteringPolicy>
 <SearchFilteringPolicy>
    <Identifier>1</Identifier>
    <Description>Filter for sentiment analysis</Description>
    <QualityAttributeRanges>1:0.0-1.0,2:0.5-1.0</QualityAttributeRanges>
    <MetadataIdentifier>metadata1</MetadataIdentifier> # optionally match with Identifier of metadata
    <TimeStart>2015-10-12 00:00:01 +0300</TimeStart> #Date range for metadata to be searched
    <TimeEnd>2015-12-12 00:00:01 +0300</TimeEnd>
    <IncludeData>false</IncludeData> # specify if data (social media data) should be included to the reply
  </SearchFilteringPolicy>
</PolicyCollection>


XML file over REST API (HTTP POST) for the creation of popularity quality attribute:

<QualityAttribute>
  <Identifier>3</Identifier>
  <Name>Popularity</Name>
  <DataSourceType>Twitter</DataSourceType>
  <Target>Twitter:tweets</Target>
  <QualityMetricIdsWeights>10-1.0</QualityAttributeIdsWeights>  # indicates weights to be associated with quality metrics for evaluation of popularity
  <QualityDataProcessingId>2</QualityDataProcessingId> # indicates the processing requirements for quality evaluation
</QualityAttribute>


XML file over REST API (HTTP POST) for the creation of quality metrics. A quality metric for evaluation of 
popularity is created. The metric is encoded in JEngineRules-format. Popularity is determined based on 
reTweetCount, numberOfFriends, and numberOfFollowers attributes of a Tweet:

<QualityMetric>
  <Identifier>10</Identifier>
  <Description>Popularity estimation based on tweets</Description>
  <Purpose>Sentiment analysis</Purpose>
  <Applicability>Tweets</Applicability>
  <ValueRange>0.0-1.0</ValueRange>
  <Format>JSR94</Format>
  <Formula>
  <![CDATA[
<?xml version="1.0" encoding="UTF-8"?>
<rule-execution-set>
   <name>RuleExecutionSet1</name>
   <description>Rule Execution Set</description>
 
   <synonymn name="prop" class="org.jruleengine.Clause" /> 

    <!--
      This rule set defined popularity condition for Twitter data.
    -->

   <!-- Range reTweetCount: 5-10 -> 0.1    -->
   <rule name="Rule1" description="popularity-retweet1" >
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getVariable1" op="=" rightTerm="reTweetCount" />
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getValue1" op="&lt;=" rightTerm="10" />
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getValue1"  op="&gt;" rightTerm="5" />
    <then method="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.addToValue" arg1="0.1" />
   </rule> 

   <!-- Range reTweetCount: 10-100 -> 0.2    -->
   <rule name="Rule2" description="popularity2-retweet2" >
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getVariable1" op="=" rightTerm="reTweetCount" />
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getValue1" op="&lt;=" rightTerm="100" />
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getValue1"  op="&gt;" rightTerm="10" />
    <then method="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.addToValue" arg1="0.2" />
   </rule> 

  <!-- Range reTweetCount: 100-x -> 0.33    -->
   <rule name="Rule3" description="popularity2-retweet4" >
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getVariable1" op="=" rightTerm="reTweetCount" />
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getValue1"  op="&gt;" rightTerm="100" />
    <then method="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.addToValue" arg1="0.33" />
   </rule> 


   <!-- Range numberOfFriends: 100-1000 -> 0.1    -->
   <rule name="Rule4" description="popularity-numberOfFriends1" >
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getVariable2" op="=" rightTerm="numberOfFriends" />
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getValue2" op="&lt;=" rightTerm="1000" />
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getValue2"  op="&gt;" rightTerm="100" />
    <then method="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.addToValue" arg1="0.1" />
   </rule> 

   <!-- Range numberOfFriends: 1000-10000 -> 0.2    -->
   <rule name="Rule5" description="popularity-numberOfFriends2" >
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getVariable2" op="=" rightTerm="numberOfFriends" />
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getValue2" op="&lt;=" rightTerm="10000" />
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getValue2"  op="&gt;" rightTerm="1000" />
    <then method="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.addToValue" arg1="0.2" />
   </rule> 

   <!-- Range numberOfFriends: 10000-x -> 0.33    -->
   <rule name="Rule6" description="popularity-numberOfFriends3" >
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getVariable2" op="=" rightTerm="numberOfFriends" />
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getValue2"  op="&gt;" rightTerm="10000" />
    <then method="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.addToValue" arg1="0.33" />
   </rule> 


   <!-- Range numberOfFollowers: 100-1000 -> 0.1    -->
   <rule name="Rule7" description="popularity-numberOfFollowers1" >
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getVariable3" op="=" rightTerm="numberOfFollowers" />
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getValue3" op="&lt;=" rightTerm="1000" />
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getValue3"  op="&gt;" rightTerm="100" />
    <then method="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.addToValue" arg1="0.1" />
   </rule> 

  <!-- Range numberOfFollowers: 1000-10000 -> 0.2    -->
   <rule name="Rule8" description="popularity-numberOfFollowers2" >
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getVariable3" op="=" rightTerm="numberOfFollowers" />
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getValue3" op="&lt;=" rightTerm="10000" />
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getValue3"  op="&gt;" rightTerm="1000" />
    <then method="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.addToValue" arg1="0.2" />
   </rule> 

  <!-- Range numberOfFollowers: 10000-x -> 0.33    -->
   <rule name="Rule9" description="popularity-numberOfFollowers3" >
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getVariable3" op="=" rightTerm="numberOfFollowers" />
    <if leftTerm="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.getValue3"  op="&gt;" rightTerm="10000" />
    <then method="metadata.qualitymanagement.qualityattributes.metricrules.PopularityCondition.addToValue" arg1="0.33" />
   </rule> 
 </rule-execution-set> 
]]>
</Formula>
</QualityMetric>


Metadata/XML file in REST API (HTTP POST) for starting of the quality evaluation process:

<Metadata>
  <Tags>tagA,tagB</Tags>
  <DatasetLocation>URL</DatasetLocation>
  <DataSource>sourceX</DataSource>
  <DataSourceType>Twitter</DataSourceType>
  <ProcFunction>SentimentAnalysisImplX</ProcFunction>
  <ProcApp>myAppX</ProcApp>
  <Meaning>Data is used for getting insight.</Meaning>
  <Purpose>Improve decision making.</Purpose>
  <Identifier>metadata1</Identifier>
  <Language>en</Language>
  <ContentDescription>tweets are saved into JSON files</ContentDescription>
  <Creator>personX in CompanyY</Creator>
  <PlaceLocation>GPS coordinates</PlaceLocation>
  <ContentType>semi-structured</ContentType>
  <ContentFormat>json</ContentFormat>
  <DataSize>5000</DataSize>
  <RequiredSW>MyProcessingAppParser</RequiredSW>
  <DecisionPoint>Sentiment analysis</DecisionPoint>
  <Timestamp>2015-12-12 00:00:01 +0300</Timestamp>
  <DecisionPointRequired>false</DecisionPointRequired>
  <DataProvider>Twitter Inc.</DataProvider>
  <Licenses>GPL</Licenses>
  <AccessRights>CompanyX</AccessRights>
  <Copyright>Twitter</Copyright>
  <DataPrivacy>public</DataPrivacy>
  <DataStorageIndicator>Store-DataItems</DataStorageIndicator> # indicates if data items (social media data) should be stored
</Metadata>


HTTP GET for searching of quality data with the specified search filtering policy:

http://location:port/rest/metadataapi/searchpolicy?Policies=1=SearchFiltering



XML content (in HTTP 200 OK) response to a end user query:

<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<Items> 
<Metadata> 
<Identifier>metadata1</Identifier> # associated with a data set 
<DataItemMetadata> 
  <Identifier>metadata1</Identifier> 
  <QuaParam1Description>Timeliness</QuaParam1Description> # quality attribute
  <QuaParam1Value>1.0</QuaParam1Value> # value of quality attribute
  <QuaParam2Description>Popularity</QuaParam2Description> # quality attribute
  <QuaParam2Value>0.4</QuaParam2Value> # value of quality attribute
  <DataItemId>760041876305145856</DataItemId> # tweet identifier 
  <Data>tweet data</Data>
 </DataItemMetadata>
</Metadata> 
</Items>