13

I was wondering if anybody knew of any good Perl modules and/or Java classes for sentiment analysis. I have read about LingPipe, but the program would eventually need to be used for commercial use so something open-source would be better. I also looked into GATE, but their documentation on sentiment analysis is sparse at best.

Fred Foo
  • 355,277
  • 75
  • 744
  • 836
user387049
  • 6,647
  • 8
  • 53
  • 55

3 Answers3

7

Have a look at Rate_Sentiment in the WebService::GoogleHack module at CPAN. There's more information about the project at SourceForge.

Pedro Silva
  • 4,672
  • 1
  • 19
  • 31
  • 1
    I saw this, but it doesn't have any ratings and I was worried about scalability since it mentions using the Google API. Do you have any experience with it yourself? – user387049 Nov 15 '10 at 19:10
  • No, I've never used it. However, [Ted Pedersen](http://www.d.umn.edu/~tpederse/) seems to be one of the authors. He's a [pretty major figure](http://search.cpan.org/~tpederse/) in Perl NLP. – Pedro Silva Nov 15 '10 at 19:14
3

I just added a sentiment analysis library to my Social Media Analytics Research Toolkit. The blog post / announcement is here. It's in R, not in Java, but there's a good interface between R and Java in the toolkit, so you can write your "glue code" in Java to call the R library. There's also an R - Python interface in the toolkit.

There's supposed to be an R / Perl interface too, but I haven't been able to contact the maintainer about bugs, so I took it out of the build.

znmeb
  • 31
  • 1
0

You might want to take a look at LingPipe (Java) based sentiment analysis at:

http://alias-i.com/lingpipe/demos/tutorial/sentiment/read-me.html

and GATE (http://gate.ac.uk/sentiment/)

For more generalized NLP parsers see The Stanford parser (http://nlp.stanford.edu/software/lex-parser.shtml), NLTK (Python) (http://www.nltk.org/), etc.

I'm not aware of any similar open source tools for Perl, although there are some good basic references out there to get you started, e.g.:

Billisoly, R. (2008) Practical Text Mining with Perl. Wiley. ISBN 978-0-470-17643-6.

Jonathan Hujsak
  • 69
  • 3