Options
Information Extraction Plugin and may be coreference-chains
Contributor I
Dear Rapidminer,
I am a really new user of RapidMiner and I am searching for the “IE-plugin”, which is described in the paper “Information Extraction with RapidMiner ” from Felix Jungermann. I’ve installed the “Text Processing” – and the “Weka”- plugin but I have not found the CRF-Operator or anything to annotated manually entities in the sample texts.
Moreover I’m interested to calculate coreference-chains. So is there any module in the rapidminer-framework, which does the same like the following GATE-modules :
Named Entity Transducer
OrthoMatcher
Pronominal Coreference Module
???
Thanks for any help,
greetings,
Isabel
I am a really new user of RapidMiner and I am searching for the “IE-plugin”, which is described in the paper “Information Extraction with RapidMiner ” from Felix Jungermann. I’ve installed the “Text Processing” – and the “Weka”- plugin but I have not found the CRF-Operator or anything to annotated manually entities in the sample texts.
Moreover I’m interested to calculate coreference-chains. So is there any module in the rapidminer-framework, which does the same like the following GATE-modules :
Named Entity Transducer
OrthoMatcher
Pronominal Coreference Module
???
Thanks for any help,
greetings,
Isabel
0
Answers
probably the right person to ask this question is Felix :-) As far as I know the plugin is not published yet, but Felix will present it at the RComm 2010 (http://www.rcomm2010.org/).
Best,
Simon
I ported the Information Extraction plugin - I never really published up to now - to RapidMiner 5!
If you are interested in testing the preliminary version which is under permanent development you might download it at:
http://www-ai.cs.tu-dortmund.de/SOFTWARE/IEPLUGIN
But unfortunately, coreference-chains are not supported up to now!
Best,
Felix
I am also highly interested in the new extension, I was going to develop something similar on my own (as part of my thesis). But it would be nonsense to implement an existing extension again, so I could perhaps shift my focus to some other problems. I know, the extension will first be released during RCOMM, but is there some information material already available? I downloaded the current development version but I am not sure how to design processes correctly. I don't even know how to bring up the TextAnnotator GUI. Is this possible at all without further changes to the RapidMiner core? I would appreciate some hints, otherwise I will exercise patience and wait until the extension is officially released.
Regards,
Matthias
the plugin now is published on sourceforge: http://sourceforge.net/projects/ieplugin4rm/
The next step will be to offer some exemplary processes.
Additionally, I am planning to develop a users guide.
Best,
Felix