Sarcasm Detection Model For Afaan Oromo Text Using Machine Learning Approach

Abstract

Sarcasm refers to the use of words that mean the opposite of what you really want to say, especially in order to insult someone, or to show irritation, or just to be funny. People frequently convey it vocally by using strong tonal stress and certain nonverbal indications such as eye rolling. Which is obviously not available for expressing sarcasm in text? This is a crucial step to sentiment analysis, considering the prevalence and challenges of sarcasm in sentiment-bearing text. Sarcasm detection is the task of predicting sarcasm in text Therefore, in this thesis we developed a model to detect the presence of sarcasm in Afaan Oromo texts. We used primary data‟s from BBC Afaan Oromo and wirtuu jildii 8ffaa published by Oromo culture Centre We used lexical (unigram), Emoticons (smiley faces etc) and Semantic features to extract different feature sets as useable inputs for Machine learning. such as Support Vector Machine (SVM), multinomial Naïve Bayes, binomial Naïve Bayes, logistic Regression, Random forest, k-nearest neighbour (knn), decision tree, adaboost and gradient boost classifiers with two selected feature extraction models tf-idf and BOW are used. all models is tested with difference evaluation metrics among these Support Vector Machine (SVM) with TF-IDF got better accuracy of 93% for performance of this developed models of sarcasm detection. At all we found some strong features that characterize sarcastic texts. However, a machine learning based features proved more promising in identifying the various facets of sarcasm