Hello everyone, recently I've posted a message related to extending KIM ontology and using protont:Topic class. I did everything according the responses I received: extended the ontology, made a knowledge base populated with domain specific instances (use the wkb.nt namespace, defined all instances as generatedBy an instance of the TrustedSource, added aliases to them) and changed %KIM_HOME%/config/sesame.inmem.conf file to include reference to the ontology extension and the knowledge base. However, when I tried to annotate same sample text (using the code from the examples provided in the documentation) it did not recognize any domain specific concept! I even noticed that it did not recognize (and annotate) some instances from wkb.nt knowledge base - for example I tried to annotate a text about some sport events, but even though the words "golf" and "rugby" were present in the text, KIM did not recognize them as instances of the class Sport (as I supposed it would do, since they are defined as aliases of instances of the Sport class in wbk.nt).
Can you please tell me where did I go wrong? And what is to be additionally done to enable KIM recognize domain concepts? If I have to extend the Information Extraction module, can you just give me some hints how to do that? Any kind of help would be more than welcomed:-) Jelena _______________________________________________ NOTE: Please REPLY TO ALL to ensure that your reply reaches all members of this mailing list. KIM-discussion mailing list [email protected] http://ontotext.com/mailman/listinfo/kim-discussion_ontotext.com
