From genesisonto at gmail.com Thu Jan 5 21:33:41 2012 From: genesisonto at gmail.com (Onto Genesis) Date: Thu, 5 Jan 2012 21:33:41 -0500 Subject: [Kim-discussion] Using RelFinder with KIM Message-ID: I was wondering if RelFinder can be configured with KIM. RelFinder information can be found here http://www.visualdataweb.org/relfinder.php One needs to specify a URI that enables SPARQL access to the RDF dataset. The intent is to be able to use RelFinder to query or discover relations between entities that are extracted and stored in OWLIM instance that comes with KIM. Thanks -- Genesis From reneta.popova at ontotext.com Thu Jan 12 06:18:01 2012 From: reneta.popova at ontotext.com (Reneta Popova) Date: Thu, 12 Jan 2012 13:18:01 +0200 Subject: [Kim-discussion] REGULAR REGISTRATION CLOSING SOON! Semantic Technologies with OWLIM, 25-27 Jan, London Message-ID: <6337ECDC-8985-4683-8464-BFA96AD3AD2B@ontotext.com> Dear Ontotext friends, Regular registration closing soon! Book your place now for Semantic Technologies with OWLIM! After the 15th of January incur a 30% surcharge (only 20% surcharge for students and partners). The number of attendants is limited, as the maximum course size is 12 and there are already several people registered. The Semantic Technologies with OWLIM training course is the fourth of its kind, after two successfully conducted trainings in UK and one in US. It will be held 25-27th January 2012 in Central London. We?re very excited to bring you a full day of introduction in Semantic Technologies, followed by two full days of tech specifics about OWLIM. Semantic Technologies with OWLIM focuses on: the basic principles of semantic technology, W3C standards (RDF, OWL, SPARQL), knowledge representation, reasoning and linked data building applications based on OWLIM, including advanced features such as owl:sameAs optimisation, geo-spatial and full-text search indexing supporting applications based on OWLIM, including cluster configuration and maintenance. Participants will be able to select only those days suitable for their level of experience.The trainers have many years of experience in teaching semantic technology and involvement in landmark deployments of RDF database engines, like the one for BBC's World Cup website, and in outstanding linked data projects like LinkedBrainz. Full details are available from here. The Ontotext team -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: tbg.jpg Type: image/jpg Size: 2744 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: top.jpg Type: image/jpg Size: 7283 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: bottom.jpg Type: image/jpg Size: 18308 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: bbg.jpg Type: image/jpg Size: 2785 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: bg_pattern.jpg Type: image/jpg Size: 8416 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: lbg.jpg Type: image/jpg Size: 1330 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: rbg.jpg Type: image/jpg Size: 1319 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: letter_bg.jpg Type: image/jpg Size: 1381 bytes Desc: not available URL: From sreckojoksimovic at gmail.com Sat Jan 21 19:04:53 2012 From: sreckojoksimovic at gmail.com (Srecko Joksimovic) Date: Sun, 22 Jan 2012 01:04:53 +0100 Subject: [Kim-discussion] Annotate Unicode text Message-ID: <004a01ccd899$788cefc0$69a6cf40$@com> Hi, Is it possible to annotate cirillyc text using KIM platform? I have read this publication [GeorgievEtAl2011] Georgi Georgiev, Valentin Zhikov, Borislav Popov, and Preslav Nakov. Building a Named Entity Recognizer in Three Days: Application to Disease Name Recognition in Bulgarian Epicrises. In Proceedings of the RANLP'2011 Workshop on Biomedical Natural Language Processing ( BiomedicalNLP'11). The paper. Could you please explain to me if something like that is possible using cirillyc text? Best, Srecko -------------- next part -------------- An HTML attachment was scrubbed... URL: From philip.alexiev at ontotext.com Mon Jan 23 05:04:15 2012 From: philip.alexiev at ontotext.com (Philip Alexiev) Date: Mon, 23 Jan 2012 12:04:15 +0200 Subject: [Kim-discussion] Annotate Unicode text In-Reply-To: <004a01ccd899$788cefc0$69a6cf40$@com> References: <004a01ccd899$788cefc0$69a6cf40$@com> Message-ID: <8B45F39B-93CE-43FA-A317-255A9B06E8E5@ontotext.com> Hello Srecko, The short answer is - it is possible, but not directly out of the box. The paper you noted addresses exactly this problem. In order to be able to recognize named entities in cyrillic, you need to use resources, tuned for the target language. The paper you mentioned: http://www.aclweb.org/anthology/W/W11/W11-4205.pdf and also this paper: http://www.dcs.shef.ac.uk/intranet/research/resmes/CS0201.pdf which is created by the GATE team and is dedicated to "SLAVONIC NAMED ENTITIES IN GATE" may serve as starting point and reference for future exploration of the task. The task is not be easy, as there are not many specialized resources and it will require a lot of manual work. But the above documents provide a good roadmap. As KIM uses GATE internally as a module to create semantic annotations, everything in these papers is valid and directly applicable. The role of KIM is to annotate, using GATE, then to create indexes, based on the annotations. Hope this helps, Philip On 22 Jan 2012, at 2:04 AM, Srecko Joksimovic wrote: > Hi, > > Is it possible to annotate cirillyc text using KIM platform? I have read this publication > > [GeorgievEtAl2011] Georgi Georgiev, Valentin Zhikov, Borislav Popov, and Preslav Nakov. Building a Named Entity Recognizer in Three Days: Application to Disease Name Recognition in Bulgarian Epicrises. In Proceedings of the RANLP'2011 Workshop on Biomedical Natural Language Processing (BiomedicalNLP'11). The paper. > > Could you please explain to me if something like that is possible using cirillyc text? > > Best, > Srecko > _______________________________________________ > Kim-discussion mailing list > Kim-discussion at ontotext.com > http://ontotext.com/mailman/listinfo/kim-discussion -------------- next part -------------- An HTML attachment was scrubbed... URL: From sreckojoksimovic at gmail.com Mon Jan 23 06:43:43 2012 From: sreckojoksimovic at gmail.com (srecko joksimovic) Date: Mon, 23 Jan 2012 12:43:43 +0100 Subject: [Kim-discussion] Annotate Unicode text In-Reply-To: <8B45F39B-93CE-43FA-A317-255A9B06E8E5@ontotext.com> References: <004a01ccd899$788cefc0$69a6cf40$@com> <8B45F39B-93CE-43FA-A317-255A9B06E8E5@ontotext.com> Message-ID: Thank you very much Philip! I will look this in more details. It is very hard, but I will try something. What is the case with Latin? Serbian, Croatian Latin? Best, Srecko On Mon, Jan 23, 2012 at 11:04 AM, Philip Alexiev < philip.alexiev at ontotext.com> wrote: > Hello Srecko, > > The short answer is - it is possible, but not directly out of the box. > > The paper you noted addresses exactly this problem. In order to be able > to recognize named entities in cyrillic, you need to use resources, tuned > for the target language. > > The paper you mentioned: > http://www.aclweb.org/anthology/W/W11/W11-4205.pdf > > and also this paper: > http://www.dcs.shef.ac.uk/intranet/research/resmes/CS0201.pdf > > which is created by the GATE team and is dedicated to "SLAVONIC NAMED > ENTITIES IN GATE" may serve as starting point and reference for future > exploration of the task. The task is not be easy, as there are not many > specialized resources and it will require a lot of manual work. But the > above documents provide a good roadmap. > > As KIM uses GATE internally as a module to create semantic annotations, > everything in these papers is valid and directly applicable. The role of > KIM is to annotate, using GATE, then to create indexes, based on the > annotations. > > Hope this helps, > Philip > > On 22 Jan 2012, at 2:04 AM, Srecko Joksimovic wrote: > > Hi,**** > ** ** > Is it possible to annotate cirillyc text using KIM platform? I have read > this publication**** > ** ** > [GeorgievEtAl2011] Georgi Georgiev, Valentin Zhikov, Borislav Popov, > and Preslav Nakov. *Building a Named Entity Recognizer in Three Days: > Application to Disease Name Recognition in Bulgarian Epicrises*. In > Proceedings of the RANLP'2011 Workshop on Biomedical Natural Language > Processing (BiomedicalNLP'11 ). > The paper > .**** > ** ** > Could you please explain to me if something like that is possible using cirillyc > text?**** > ** ** > Best,**** > Srecko**** > _______________________________________________ > Kim-discussion mailing list > Kim-discussion at ontotext.com > http://ontotext.com/mailman/listinfo/kim-discussion > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: