Hello,
I am an IT Specialist for a center for applied social sciences at a
university in Oklahoma. The researchers I work for have several
requests for data and data manipulation that appear to only be
solvable by NLP techniques. My background is in Computer Science, but
I have no experience with NLP other than what I've recently learned,
which isn't much.
I have two problems to solve:
1. Search through a database of nouns, verbs, and modifiers and find
all the similarly related words and measure how close the relationship
is. I believe that Wordnet::Similarity will solve this to a high
degree of accuracy.
2. Search through academic vitae, extract important named entities
such as conferences attended, publications, names, and locations.
I've looked at the Python NLTK and several of the Perl Lingua modules
for this, but am unsure of the optimal strategy. I would prefer
advice on what areas I should study and your favorite books for
learning this material. My current uneducated strategy is this. Use
a POS tagger on the vitae. Implement algorithms for named entity
extraction to get the name and university/company of the vitae. Try
to guess the publications and journal section of the vitae. Run the
named entities against a journal and conference dictionary, then
assume the other named entities in the journal and conference sections
are either locations or names of associates. I'm sure my plan is
quite lacking in many ways, which is why I would also love to hear
what your favorite "How I learned NLP" books. And unfortunately, the
scope of my project does not allow enough time for my acquiring a
Master's in linguistics.
This yahoo group appears to be more research oriented, so if my
request is out of place, please let me know.
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/nlpatumd/
<*> To unsubscribe from this group, send an email to:
[EMAIL PROTECTED]
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
