I am learning sentiment analysis to apply it to twitter real time data to predict user's mood. I ponder about using which alternative way to do that data mining job.
- Use all words to process and create a model by SVM or such classifier
- Get vast lexicon list that is obtained form a outsource and use it as a vocabulary and just use these lexicons as a features on classifier.
- Just use lexicons and instead of using a model emerged by a classifier, take the real time value and look at frequencies of lexicons at the tweets and rank. (this is not the way for me but I want to see the public wisdom about)
- Or use any other method.
As a sub question, If predicting mood will be complicated for me I will just take care of the binary classification as positive or negative feelings. Which method will work best in that case?
