DUTH at SemEval-2021 Task 7: Is Conventional Machine Learning for Humorous and Offensive Tasks enough in 2021?

Proceedings of the 15th International Workshop on Semantic Evaluation (SemEval-2021), pages 1125–1129 Bangkok, Thailand (online), August 5–6, 2021. ©2021 Association for Computational Linguistics 1125 DUTH at SemEval-2021 Task 7: Is Conventional Machine Learning for Humorous and Offensive Tasks enough in 2021? Alexandros Karasakalidis Dimitrios Effrosynidis Avi Arampatzis Database & Information Retrieval research unit, Department of Electrical & Computer Engineering, Democritus University of Thrace, Xanthi 67100, Greece {alexkara23,deffrosy,avi}@ee.duth.gr Abstract This paper describes the approach that was developed for SemEval 2021 Task 7 Ha- hackathon: Incorporating Demographic Fac- tors into Shared Humor Tasks (Meaney et al., 2021) by the DUTH Team. We used and com- pared a variety of preprocessing techniques, vectorization methods, and numerous conven- tional machine learning algorithms, in order to construct classification and regression mod- els for the given tasks. We used majority vot- ing to combine the models’ outputs with small Neural Networks (NN) for classification tasks and their mean for regression for improving our system’s performance. While these meth- ods proved weaker than modern, deep learning models, they are still relevant in research tasks because of their low requirements on computa- tional power and faster training. 1 Introduction The underpinnings of humor have proven far more vexing than those of other emotional experiences. It is a highly subjective topic that various schol- ars have attempted to construct theories for under- standing its fundamental elements in the studies of philosophy, linguistics, psychology and soci- ology. Some theories, e.g. the Benign Violation Theory (Warren and McGraw, 2015), suggest that humor can be described as linguistic violations that still make grammatical sense. The aforementioned theory supports that for a joke to be classified as humorous, it needs to avoid being too harmless or too offensive. There are numerous studies in humor sentiment analysis in the last decade. In microblogging, Reyes et al. (2012) considered extracting linguistic devices from tweets to be used as features for clas- sifying these tweets as humorous or ironic, while Raz (2012) approached the classification of humor- ous tweets as a multi-class problem of 11 types of humor, in his attempts to better attribute the real sentiment of a tweet. Recent attempts on humor detection on SemEval’s 2020 Task 7 indicate that transformer models like BERT (Mahurkar and Patil, 2020) far outperform traditional machine learning algorithms (S et al., 2020). This paper describes our submissions to Se- mEval 2021 task 7 and is structured as follows. Section 2 describes the tasks, training data, and evaluation measures. Section 3 describes key meth- ods and algorithms used. Section 4 describes our proposed system while Section 5 analyzes our re- sults. Finally, we draw our conclusions in Section 6, where we also propose directions for future work. 2 Background In this section we describe each subtask’s objective, the given data, and evaluation measures. 2.1 Subtasks The main objectives of SemEval 2021 Task 7 were split into 4 subtasks. Subtask 1a required us to clas- sify short texts as humorous or not, while Subtasks 1b and 1c required us to …

Content truncated to 500 words