Besides the line "Semantic composition is accomplished using backward chaining and other AI planning algorithms." in the alphaWorks site, I could not find any documentation on the approach for composition used by the "Semantic Tools for Web Services" Eclipse plug-ins.
Are there any papers published that describe the approach?
Is there any documentation to help interpret the compositions obtained? (Which seem to me a sequence of unconnected service invocations, see below one of the composition generated for the sample "textanalysis-req-sem.wsdl" in this wiki)
kind regards,
João Paulo Almeida
----
<process xmlns="http://schemas.xmlsoap.org/ws/2003/03/business-process/"
name="SomeName">
<partnerLinks>
</partnerLinks>
<partners>
</partners>
<variables>
</variables>
<sequence name="SomeSequence">
<invoke name="invoke-{http://uima.semantica.ibm.com.wsdl/tokenizer-v1/}tokenizer/tokenizer_OBJECT"
operation="{http://uima.semantica.ibm.com.wsdl/tokenizer-v1/}tokenizer/tokenizer"
>
</invoke>
<invoke name="invoke-{http://uima.semantica.ibm.com.wsdl/tokenizer-v1/}tokenizer/tokenizer_OBJECT"
operation="{http://uima.semantica.ibm.com.wsdl/tokenizer-v1/}tokenizer/tokenizer"
>
</invoke>
<invoke name="invoke-{http://uima.semantica.ibm.com.wsdl/lexicalanalyzer-v1/}lexicalanalyzer/lexicalanalyzer_OBJECT"
operation="{http://uima.semantica.ibm.com.wsdl/lexicalanalyzer-v1/}lexicalanalyzer/lexicalanalyzer"
>
</invoke>
<invoke name="invoke-{http://uima.semantica.ibm.com.wsdl/lexicalanalyzer-v1/}lexicalanalyzer/lexicalanalyzer_OBJECT"
operation="{http://uima.semantica.ibm.com.wsdl/lexicalanalyzer-v1/}lexicalanalyzer/lexicalanalyzer"
>
</invoke>
<invoke name="invoke-{http://uima.semantica.ibm.com.wsdl/namerecognizer/}namerecognizer/namerecognizer_OBJECT"
operation="{http://uima.semantica.ibm.com.wsdl/namerecognizer/}namerecognizer/namerecognizer"
>
</invoke>
</sequence>
</process>
