InformationExtractor

InformationExtractor extracts entities such as names and addresses from strings of unstructured data (also known as plain text).

It is possible that not all entities for any selected type will be returned because accuracy varies depending on the type of input. Because Information Extractor uses natural-language processing, a string containing a grammatically correct sentence from a news article or blog would likely have a more accurate return of names than a simple list of names and dates.

Resource URL

http://server:port/soap/InformationExtractor

Example

This shows a SOAP request:

<soapenv:Envelope xmlns:soapenv="http://schemas.xmlsoap.org/soap/envelope/" xmlns:inf="http://www.pb.com/spectrum/services/InformationExtractor">
   <soapenv:Header/>
   <soapenv:Body>
      <inf:InformationExtractorRequest>
         <inf:options>
            <inf:EntityList>Person</inf:EntityList>
         </inf:options>
         <inf:input_port>
            <inf:PlainText>
               <inf:PlainText>My name is Arthur Pitney</inf:PlainText>
            </inf:PlainText>
            <inf:PlainText>
               <inf:PlainText>My name is Walter Bowes</inf:PlainText>
            </inf:PlainText>
         </inf:input_port>
      </inf:InformationExtractorRequest>
   </soapenv:Body>
</soapenv:Envelope>

This would be the response:

<soap:Envelope xmlns:soap="http://schemas.xmlsoap.org/soap/envelope/">
   <soap:Body>
      <ns3:InformationExtractorResponse xmlns:ns2="http://spectrum.pb.com/"​xmlns:ns3="http://www.pb.com/spectrum/services/InformationExtractor">
         <ns3:output_port>
            <ns3:Result>
               <ns3:Entity>
                  <ns3:Entity>
                     <ns3:Text>Arthur Pitney</ns3:Text>
                     <ns3:Type>Person</ns3:Type>
                  </ns3:Entity>
               </ns3:Entity>
               <ns3:user_fields/>
            </ns3:Result>
            <ns3:Result>
               <ns3:Entity>
                  <ns3:Entity>
                     <ns3:Text>Walter Bowes</ns3:Text>
                     <ns3:Type>Person</ns3:Type>
                  </ns3:Entity>
               </ns3:Entity>
               <ns3:user_fields/>
            </ns3:Result>
         </ns3:output_port>
      </ns3:InformationExtractorResponse>
   </soap:Body>
</soap:Envelope>