Skip to content

Latest commit

 

History

History
25 lines (15 loc) · 992 Bytes

SentimentAnalysis.md

File metadata and controls

25 lines (15 loc) · 992 Bytes

I love beer (["I" "PRP"] ["love" "VBP"] ["beer" "NN"])

I do like beer (["I" "PRP"] ["do" "VBP"] ["like" "IN"] ["beer" "NN"])

I don't like beer (["I" "PRP"] ["do" "VBP"] ["n't" "RB"] ["like" "VB"] ["beer" "NN"])

I do not like beer (["I" "PRP"] ["do" "VBP"] ["not" "RB"] ["like" "VB"] ["beer" "NN"])

The movie was bad (["The" "DT"] ["movie" "NN"] ["was" "VBD"] ["bad" "JJ"])

The movie sucked. (["The" "DT"] ["movie" "NN"] ["sucked" "VBD"] ["." "."])

===

Some tipps: http://scialert.net/fulltext/?doi=tasr.2011.1141.1157&org=10

Negation: Negation words reverse the polarity of opinion words by checking their position in a sentence. The words are (Not, Never, N’t, Doesn’t, Can’t, Nor, Don’t, Wouldn’t, No, etc). The result will be the opposite if the system fails to recognize the negation word. So, for the recognition of the semantic expression in the sentence, we use the word sense disambiguation to extract the exact or nearest semantic score of the opinion expression