The ExtractSentiment function extracts the sentiment (positive, negative, or neutral) of each input document or sentence, using either a classification model output by the function TrainSentimentExtractor or a dictionary model.
The dictionary model consists of WordNet, a lexical database of the English language, and the following negation words:
- no
- not
- neither
- never
- scarcely
- hardly
- nor
- little
- nothing
- seldom
- few
The function handles negated sentiments as follows:
- -1 if the sentiment is negated (for example, "I am not happy")
- -1 if the sentiment and a negation word are separated by one word (for example, "I am not very happy")
- +1 if the sentiment and a negation word are separated by two or more words (for example, "I am not saying I am happy")
This function can be used with real-time applications. See AMLGenerator.