Code example for Sentiment Analysis for Asian languages - Python NLTK

Question

There is a demo on sentiment analysis with NLTK (python) here http://text-processing.com/demo/sentiment/.

And also the tutorials on the parts of sentiment analysis

• http://streamhacker.com/2010/06/16/text-classification-sentiment-analysis-eliminate-low-information-features/
• http://streamhacker.com/2010/05/10/text-classification-sentiment-analysis-naive-bayes-classifier/
• http://nltk.googlecode.com/svn/trunk/doc/book/ch07.html

Is there any full code example or working projects with python NLTK on sentiment analysis for Asian languages? (especially for Chinese, Japanese, Korean or Arabic, Hebrew and Persian languages)

Quizzical

hi, did you find a solution for this problem to support many languages? – Echidna

Answer 1

In the Japanese version of Natural Language Processing with Python there is an extra-chapter dedicated to Japanese processing: https://nltk.googlecode.com/svn/trunk/doc/book-jp/ch12.html (in Japanese).

Unless you're looking for symbolic approaches to sentiment analysis, you can apply existing technics to Japanese (or whatever language you want to process) and use language specific tools for the preprocessing step.

Surd

Answer 2

I don't see any examples using NLTK, but the chatterbox (social media oreinted) and chinese-analytics (business oriented) libraries can be used to perform sentiment analysis for Chinese languages.

Khabarovsk