Humor is an irresistible component of daily human interactions and communication. Owing to its wide range of uses and increasing popularity on websites for social networking, it has become crucial for computers to be able to comprehend humor. In order to facilitate the computational analysis of humor, plan to identify humor via machine learning. Social networking platforms like Facebook, Twitter, and Reedit have plenty of humor. As a result, the authors utilized their dataset to conduct this research. In this paper, a hybrid machine-learning model for detecting humor in English sentences (texts) is investigated. The proposed model takes into account word n-gram TF-IDF characteristics. At last, the authors found that the proposed model obtained a macro F1 score of 0.8103 and an accuracy of 0.8223. The proposed model’s source code ( https://github.com/Shweta-Thyro/Humor_Detection_Hybrid_Model ) is available.