Advanced Sarcasm Detection in Newspaper Headlines: A Comparative Study of Machine Learning Approaches

This study presents a comprehensive analysis of sarcasm detection in newspaper headlines using various machine learning techniques. We explore the effectiveness of logistic regression, Naïve Bayes (multinomial and Gaussian), random forest, and support vector machine (SVM) models in identifying sarcastic content within the concise format of news titles. Our research addresses the unique challenges posed by the brevity and impactful language characteristic of headlines. The study utilizes a curated dataset of sarcastic and non-sarcastic headlines, employing natural language processing techniques for preprocessing and feature extraction. Performance evaluation metrics include accuracy, precision, recall, and F1-score. Results indicate that the random forest model outperforms other approaches, achieving 94% accuracy in sarcasm detection. This research contributes to the growing field of sentiment analysis and offers insights into the nuanced task of decoding sarcasm in condensed textual formats. KEYWORDS: Random Forest, Support Vector Machine (SVM), TF-IDF, Natural Language Processing (NLP), Subtle Sarcasm