Thanks for the note and the patch. We really want the tutorials to work as written, so any suggestions for patches are greatly appreciated. I'm in the final...
LingPipe 3.8.0 is now available from the site: http://alias-i.com/lingpipe This is a big release — the details are available on the home page linked above. ...
I accidentally left an old description in the documentation on the LingPipe home page. We have not in fact deprecated util.FeatureExtractor as we said on the...
... Why do you want to do that? ... You can actually put the stemming inline using our tokenizers. Or you can preconvert it using whatever tool you want and...
Hi, Bob I was going to reply to your email on this morning with more questions, but luckily I could get back from work, study a little more about this subject...
Re-reading my emails I noticed that I didn't tell you all what exactly my final assignment is. I'll write a service to classify texts into predefined ...
... The answer is probably "yes". In fact our evals show what the effect of various thresholds are on the one-versus-all subproblems if you look at the scored...
An anonymous bug reporter indicated the model: $LINGPIPE/demos/models/langid-leipzig.classifier was missing. I didn't realize it was linked from the tutorial...
... What I didn't mention was that it's compiled with 5-grams and then pruned to a minimum count per n-gram of 10. That resulted in a 10MB model that's not...
I'm interested in sentimental analysis and started few tests of detection. Is it possible do sentimental analysis comparatively? For example: "Java is better...
Hello, My question is similar to Gustavo's (and I closely looked at the sentiment detection tutorial a looong time ago). How well does LingPipe detect...
... I'm afraid there's nothing built into LingPipe that'll let you do this kind of comparative sentiment extraction out of the box. But there are some...
We just released a patch for LingPipe 3.8. As usual, it's available from the home page: http://alias-i.com/lingpipe There are only two changes: 1. Bug...
We received a bug report that the SVD demo program in $LINGPIPE/demos/tutorial/svd/src/Lsi.java throws a null pointer exception as of LingPipe 3.8. I...
Dear all, could somebody help me with the code below. I want to extract author list from Medline citation. thanks in advance for any suggestions. Cheers,...
Dear All I've just start learning LingPipe for my project. My project focus on Sentiment Analysis. Now I want to try using new set of movie review from...
... You're on the right track -- you just need to change the handle() method to handle ciations and then extract the author list from the citation. public void...
... Starting from our sentiment tutorial, you only need to rewrite the bit that pulls text out of the file-based representations. Alternatively, you can just...
Thanks for the bug report: ... ... It should indeed be "length + start". I patched it for the next release. This is what I get for starting the API with...
I have a training set of 45K documents for 1000 categories and don't want users of my application to have to wait while the classifier is trained every time...
... I have no idea of how Daffodil will deal with java serialization but you can serialize with the AbstractExternalizable class for any of the classifiers...
... This will depend on which classifier you use. All of them either implement java.io.Serializable, or they implement com.aliasi.util.Compilable. The...
Thanks Bob, the storing of the classifier into a blob field works great. I am having a little trouble reading the object back out of the database though. I...
I fiddled with the code some more and got it to work. here is the working code... note : this is extremely fast especially since the blob created from the...
I love it when people solve their own problems before I wake up in the morning! A good way to figure out what you've got serialized is to do this: Object obj =...
Mike just found (and patched) a bug in our scored precision recall evalaution implementation. It was using a TreeSet with a score-based comparator instead of a...