GitHub - deepakkumar-b/NLP_Sarcasm_Detection: This project is an experiment to find if a tweet is sarcastic and Brown clustering of similar words, topics association with sarcasm - used to generate feature vectors, POS - tagging, n-gram feature vector generation, sentiment analysis of the tweets and scoring based on positivity(negativity) of the words in a tweet to find the mix of positive sentiment and negative actions are some of the techniques used to identify sarcasm. The 'kitchen sink' of a mix of these features helped us achieve close to ~58% accuracy. Languages

deepakkumar-b / NLP_Sarcasm_Detection Public

Notifications You must be signed in to change notification settings
Fork 1
Star 2

This project is an experiment to find if a tweet is sarcastic and Brown clustering of similar words, topics association with sarcasm - used to generate feature vectors, POS - tagging, n-gram feature vector generation, sentiment analysis of the tweets and scoring based on positivity(negativity) of the words in a tweet to find the mix of positive …

2 stars 1 fork Branches Tags Activity

Star

Notifications

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 2 Commits
.gitattributes		.gitattributes
Brown_Clustering.ipynb		Brown_Clustering.ipynb
Weka_Brown.txt		Weka_Brown.txt
Weka_with_Topic.txt		Weka_with_Topic.txt
normal_tweets.csv		normal_tweets.csv
paths_brown.txt		paths_brown.txt
polarity_twitDB_regular_test.csv		polarity_twitDB_regular_test.csv
polarity_twitDB_regular_train.csv		polarity_twitDB_regular_train.csv
polarity_twitDB_sarcasm_test.csv		polarity_twitDB_sarcasm_test.csv
polarity_twitDB_sarcasm_train.csv		polarity_twitDB_sarcasm_train.csv
sarcastic_tweets.csv		sarcastic_tweets.csv
test_sarcasm_twitdb_brown.arff		test_sarcasm_twitdb_brown.arff
test_sarcasm_twitdb_brown_topic.arff		test_sarcasm_twitdb_brown_topic.arff
topics.txt		topics.txt
train_sarcasm_twitdb_brown.arff		train_sarcasm_twitdb_brown.arff
train_sarcasm_twitdb_brown_topic.arff		train_sarcasm_twitdb_brown_topic.arff