Sentimental analysis phrase case

Good afternoon, I am trying to write a sentimental analysis application in python (using the naive classifier) ​​to categorize phrases from the news as positive or negative. And I have problems finding the right case for this. I tried using "General Inquirer" (http://www.wjh.harvard.edu/~inquirer/homecat.htm), which works fine, but I have one big problem. Since this is a list of words, not a list of phrases, I see the following problem when trying to label the following sentence:

He is not expected to win.

This sentence is classified as positive, which is incorrect. The reason for this is that “victory” is positive, but “not” has no meaning, since “not winning” is a phrase. Can anyone suggest either a corps or work around this problem? Your help and understanding is greatly appreciated.

+5
source share
2 answers

See for example: “What is cool and what’s not: learn to classify the sphere of negation to improve the analysis of feelings” in the Council, MacDonald and Velikovich

http://dl.acm.org/citation.cfm?id=1858959.1858969

and subsequent actions

http://scholar.google.com/scholar?cites=3029019835762139237&as_sdt=5,33&sciodt=0,33&hl=en

eg. Morante et al. 2011

http://eprints.pascal-network.org/archive/00007634/

+4

, , . , POS ( ) . , , , .

+3

All Articles