Multi-Rule Based Ensemble Feature Selection Model for Sarcasm Type Detection in Twitter

Algorithm 7

Modelling the user’s Mood changes.

	Input: History of tweets.
	Output: Polarity and Mood
	(1) For every sarcastic tweet, si, in Tweet set, T
	1.1. Extract the past tweets () of the user, ui
	1.2. Select the preceding and succeeding tweets
	1.3. Set si as T0
	(2) For every identified tweet in step 1.2
	2.1. Find the sentiment of every tweet (t)
	2.1.1. If sentiment score = 0, set polarity as neutral
	2.1.2. Else if sentiment score >0, set polarity as Positive
	2.1.3. Else set polarity to negative
	(3) Determine the level of mood change of the various users for varying types of sarcasm