Hi Mel, KIM comes with a good set of documentation which will help you. A good place to start is the Quick Start Guide. It is a short tour through the common process of using KIM - downloading, installing, setting up, populating, searching .
If you want to customize the information extraction process, the system documentation will be of great use: http://ontotext.com/kim/doc/KimDocs-3.0-EN/HomePage.html . Basically, you have two options for the ontology. You can replace the ontology in KIM completely with your own. Or, it may be more useful to map the new ontology to PROTON and reuse the knowledge already in the system. The extent of the mapping depends on how much of your ontology may be covered by PROTON - KIM's default ontology (http://proton.semanticweb.org/) . After this generally you would want to populate the system with some documents (using the populator tool) . Populating a document performs IE over it and stores the document and the entities mentioned in it in the semantic and document repositories. KIM's API provides functionality to access the documents and entities you have created. You can check it at (http://nmwiki.ontotext.com/kim-javadoc/index.html) About the semi-supervised learning - we haven't had such a use case, but we have the infrastructure to support it. Hope this helps Philip Alexiev Software Engineer, KIM team > > -------- Original Message -------- > Subject: Help requested > Date: Thu, 11 Nov 2010 08:43:54 -0800 (PST) > From: Melroy Rodrigues <[email protected]> > To: [email protected] > > I have downloaded KIM. However, I have a few questions. I need to extract > safety related information from certain texts documents and online. How would > I go about doing so? > > I would need to create an ontology and user the Populator to add that > ontology into KIM's backend? what next, how do I run my documents through KIM > so that the relevant information is extracted. Also I need to programmaticaly > work with the extracted entities. How would I do that. Finally is KIM's > ontology module support semi supervised learning? > > Thanks > Mel > > > > _______________________________________________ > Kim-discussion mailing list > [email protected] > http://ontotext.com/mailman/listinfo/kim-discussion
_______________________________________________ Kim-discussion mailing list [email protected] http://ontotext.com/mailman/listinfo/kim-discussion
