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

Algorithm 1

Preprocessing of raw twitter data.

	Input: Comma separated raw tweet file
	Output: Preprocessed and clean tweet file
	Create a list preprocessed_docs
	for each tweet S in T:
	create a list result
	for each word W in S:
	if W not in STOPWORDS:
	Add stem (W) to result
	end if
	if
	Tokenize W
	end if
	if W has @, /, RT, http
	Remove the word and then add to result
	end if
	end for
	preprocessed docs.Append (result)
	end for
	end