Hello Jörn, I am sorry, I seem to have missed all the (great) news about the GSOC. If new ideas are required for other students, I have wanted to add a probabilistic lemmatizer in OpenNLP for some time. As you know, the current lemmatizer is only dictionary based. There is an issue about adding rule-based one, but research has shown that a probabilistic lemmatizer works better for unknown words. There is already an open source tool which we could be based on to implement this into OpenNLP.
https://code.google.com/p/mate-tools This the algorithm. The first paper describes the general idea and the second presents the experiments in a realistic environment. http://grzegorz.chrupala.me/papers/chrupala-2006/paper.pdf http://grzegorz.chrupala.me/papers/chrupala-etal-2008a/paper.pdf In any case, I will open an issue about this. Rodrigo On Thu, Mar 5, 2015 at 8:04 PM, Joern Kottmann <kottm...@gmail.com> wrote: > Hello, > > we got already two students for those two GSOC WSD tasks. They contacted > us a while ago (see the WSD thread on this list) and set up the tasks so > they can apply for it. > > I am not sure if it makes much sense to break the WSD tasks further > down. > > Do you have something else in mind you could work on? I hope it is still > possible to set up new GSOC tasks. Let me check that. And we would also > need more mentors. > > HTH, > Jörn > > On Wed, 2015-03-04 at 10:41 +0530, Vidura Mudalige wrote: >> Hi all, >> >> I am Vidura, a third year Computer Science and Engineering undergraduate >> from University of Moratuwa. I'm very much interested in working with >> Apache OpenNLP project in GSoC 2015. >> >> I have worked in some open source projects. Also I have used Apache OpenNLP >> and Apache UIMA for some of my previous projects. Nowadays I am working in >> a open source project called WSO2 User Engagement Server.[1] >> >> I would like to resolve the issue OPENNLP-758.[2]. I cloned and >> successfully built the apache/opennlp.git.[3] I would like to know more >> details about the issue and expected deliverables. >> >> Thanks you. >> >> [1].https://github.com/wso2/product-ues/tree/dashboards-2.0 >> [2].https://issues.apache.org/jira/browse/OPENNLP-758 >> [3].https://github.com/apache/opennlp >