A Study in Practical Solutions to Sarcasm Detection

with Machine Learning and Knowledge Engineering Techniques

AUTHORS:
Chia Zheng Lin, Michal Ptaszynski, Fumito Masui, Gniewosz Leliwa, Michal Wroczynski

Journal:
Semantic Scholar

Link:
https://www.semanticscholar.org/paper/A-Study-in-Practical-Solutions-to-Sarcasm-Detection-Lin-Ptaszynski/2f44dfe678af36c014f5770d6c7d05ddcb31edd4

In this paper we tackled the problem of sarcasm detection with the use of machine learning and knowledge engineering techniques. Sarcasm detection is considered a complex and challenging task in Natural Language Processing and has been studied by various researchers in the past decade. To get a grasp on the present state of the art in sarcasm detection, we reviewed the important previous research in this field, with a focus on text-based sarcasm detection in English texts. In the proposed method, we compared various dataset preprocessing techniques on the proposed Deep Convolutional Neural Network model. As a result, the most specific, or least preprocessed dataset ranked as the one with the highest performance. However, we observed that some level of data preprocessing could become useful in the task of sarcasm detection.