Search the web
Sign In
New User? Sign Up
LingPipe
? Already a member? Sign in to Yahoo!

Yahoo! Groups Tips

Did you know...
Want your group to be featured on the Yahoo! Groups website? Add a group photo to Flickr.

Best of Y! Groups

   Check them out and nominate your group.
Having problems with message search? Fill out this form to ensure your group is one of the first to be migrated to the new message search system.

Messages

  Messages Help
Advanced
Messages 521 - 551 of 777   Oldest  |  < Older  |  Newer >  |  Newest
Messages: Simplify | Expand   (Group by Topic) Author Sort by Date ^
521
Dear Mr. Bob Carpenter, I posted since few weeks the fellowing email (see above) where I asked you for possible use of Lingpipe for sentence or string...
youcef bey
y_bey
Offline Send Email
Feb 7, 2008
11:20 pm
522
Dear Mr. Bob Carpenter, I posted since few weeks the fellowing email (see below) where I asked you for possible use of Lingpipe for sentence or string...
youcef bey
y_bey
Offline Send Email
Feb 7, 2008
11:24 pm
523
I'm afraid the short answer is no -- I haven't written the tutorial for string comparison and don't currently have any code samples to send you that aren't...
Bob Carpenter
colloquialdo...
Offline Send Email
Feb 8, 2008
10:32 pm
524
Hi - I'm working on Language Id and was trying to download the Leipzig Corpora from http://corpora.informatik.uni-leipzig.de/download.html. To the surprise...
Indus Khaitan
khaitan_i1
Offline Send Email
Feb 10, 2008
7:52 am
525
... I just tried it out and had no problem (i.e. it "works on my machine", where my machine is running Firefox 2.0.0.12 on Win x64 over a local ethernet...
Bob Carpenter
colloquialdo...
Offline Send Email
Feb 11, 2008
6:10 pm
526
Fantastic. It works now!...
Indus Khaitan
khaitan_i1
Offline Send Email
Feb 11, 2008
6:49 pm
527
Apologies to all, In our efforts to modernize out web infrastructure I broke the sandbox access to projects. The old address was cvs -d...
reckb
Offline Send Email
Feb 12, 2008
4:39 pm
528
Hello All, I have been looking over the demo for named entity recognizer. I did read and understood about Rule-Based Named Entity Detection and Exact ...
cambazz
Offline Send Email
Feb 13, 2008
10:35 pm
529
The best place to find info on building your own data and training a NE recognizer is in our "citationEntities" sandbox project. Instructions are at: ...
Bob Carpenter
colloquialdo...
Offline Send Email
Feb 14, 2008
12:13 am
530
So I have been playing with the citationEntities, and I run the gui tool to do the manual annotation. It did form some files like: <?xml version="1.0"...
cambazz
Offline Send Email
Feb 14, 2008
12:27 pm
531
Hello, When I am training a named entity annotator and If I have input like: "INTEL CORE 2 QUAD Q6700 2.66 GHZ 8 MB CACHE 1066 MHZ LGA775 Processor" intel ->...
cambazz
Offline Send Email
Feb 14, 2008
12:32 pm
532
... You should look at the doc in citationsEntities/readme.html The entities you annotate inside of is available on the command line. ... You can specify the...
reckb
Offline Send Email
Feb 14, 2008
6:44 pm
533
... Product identification "in the wild" is one of the harder named entity tasks. If you are tagging uniform data like a catalog or single website then it may...
reckb
Offline Send Email
Feb 14, 2008
6:54 pm
534
Hi, I recently use LingPipe for my research. I have texts from wine menus, and I use both rule-based and dictionary-based name entity recognizer for them. My...
°gÅʰª¸...
sys_admin.tw
Offline Send Email
Feb 15, 2008
5:57 am
535
Hello, I made multiple chunkers and both dictionary based and regexp based. And I have input coming line by line? How can I use them in a combined manner? ...
cambazz
Offline Send Email
Feb 15, 2008
10:02 am
536
... You still have to be careful with things like punctuation and capitalization (unless you case normalize the data). For instance, consider a wine the...
Bob Carpenter
colloquialdo...
Offline Send Email
Feb 15, 2008
6:52 pm
537
... I'm not sure what you intend here. If "input" is a string, what does the chunk() method do? It's not something from LingPipe. What you really need to do...
Bob Carpenter
colloquialdo...
Offline Send Email
Feb 15, 2008
7:28 pm
538
thanks I did some list processing, and now I can use chunkers in a chain. For a HMM named entity recognizer to work properly, how many lines are required and...
cambazz
Offline Send Email
Feb 16, 2008
1:33 am
539
... It depends on the task, of course. For persons, locations and organizations in newswire text, you start getting decent performance at 50K tokens and see...
Bob Carpenter
colloquialdo...
Offline Send Email
Feb 18, 2008
7:28 pm
540
... That's too bad. In English, there's a lot of information in capitalization for everything from sentence boundaries to spelling checking to proper noun...
Bob Carpenter
colloquialdo...
Offline Send Email
Feb 21, 2008
6:27 pm
541
Hello Bob, Some time ago I asked about ways for multi-label (when input may fall into more than one category) classification. Your answer was that regular...
lenat79
Offline Send Email
Feb 25, 2008
12:04 am
542
... The exclusivity shows up in two category distributions in the model, the prior and posterior: p(cat|ws) = p(ws|cat) * p(cat) / p(ws) posterior = likelihood...
Bob Carpenter
colloquialdo...
Offline Send Email
Feb 25, 2008
6:32 pm
543
Hi, I recently started evaluating the LingPipe SDK and I have a few problems and questions I build a big corpus, with around 50 text categories that represent ...
Nadav
nadavh080675
Offline Send Email
Mar 5, 2008
8:41 pm
545
... What exactly do you mean by text styles? As in writing styles like letter, e-mail, technical article, etc.? Are those styles mutually exclusive? If not,...
Bob Carpenter
colloquialdo...
Offline Send Email
Mar 5, 2008
9:12 pm
546
i will try to clarify my issue in regards to your questions. i have around 60 text categories, for each category i have between 200 - 1000 text files that...
Nadav Hashimshony
nadavh080675
Offline Send Email
Mar 6, 2008
7:04 am
547
... Never miss a thing. Make Yahoo your homepage. [Non-text portions of this message have been removed]...
Jemmy Lidger
jemmy.lidger
Offline Send Email
Mar 6, 2008
12:14 pm
548
... It's the amount of text, not number of files that determines sizing. ... OK -- that's fairly small. ... You probably want an NGramProcess classifier unless...
Bob Carpenter
colloquialdo...
Offline Send Email
Mar 6, 2008
5:31 pm
549
Hi bob, I used your advise and pruned the small ngrams, I use 3 as a counter and it solved the java heap problem. You wrote: You need to train the individual...
Nadav
nadavh080675
Offline Send Email
Mar 6, 2008
8:26 pm
550
Overall, you might want to rethink your classifier. Are the categories mutually exclusive and do they collectively cover all input types? Is there any ...
Bob Carpenter
colloquialdo...
Offline Send Email
Mar 6, 2008
9:17 pm
551
Hi Bob First I would like to put in a good word. I have worked with open source project in the past but never did I get such response to technical questions, ...
Nadav
nadavh080675
Offline Send Email
Mar 7, 2008
5:33 pm
Messages 521 - 551 of 777   Oldest  |  < Older  |  Newer >  |  Newest
Advanced
Add to My Yahoo!      XML What's This?

Copyright © 2009 Yahoo! Inc. All rights reserved.
Privacy Policy - Terms of Service - Guidelines - Help