kdehumor at semeval-2020 task 7: a neural network model for detecting funniness in dataset humicroedit
This work addresses humor detection in text for computational linguistics, but it is incremental as it applies existing neural methods to a specific dataset.
The authors tackled the problem of detecting humor in edited news headlines by developing a deep neural network model using BiLSTM and pre-trained embeddings, achieving results as part of the SemEval-2020 competition.
This paper describes our contribution to SemEval-2020 Task 7: Assessing Humor in Edited News Headlines. Here we present a method based on a deep neural network. In recent years, quite some attention has been devoted to humor production and perception. Our team KdeHumor employs recurrent neural network models including Bi-Directional LSTMs (BiLSTMs). Moreover, we utilize the state-of-the-art pre-trained sentence embedding techniques. We analyze the performance of our method and demonstrate the contribution of each component of our architecture.