NLPRL-IITBHU at SemEval-2018 Task 3: Combining Linguistic Features and Emoji pre-trained CNN for Irony Detection in Tweets

Abstract

This paper describes our participation in SemEval 2018 Task 3 on Irony Detection in Tweets. We combine linguistic features with pre-trained activations of a neural network. The CNN is trained on the emoji prediction task. We combine the two feature sets and feed them into an XGBoost Classifier for classification. Subtask-A involves classification of tweets into ironic and non-ironic instances, whereas Subtask-B involves classification of tweets into non-ironic, verbal irony, situational irony or other verbal irony. It is observed that combining features from these two different feature spaces improves our system results. We leverage the SMOTE algorithm to handle the problem of class imbalance in Subtask-B. Our final model achieves an F1-score of 0.65 and 0.47 on Subtask-A and Subtask-B respectively. Our system ranks 4 th on both tasks, respectively, outperforming the baseline by 6% on Subtask-A and 14% on Subtask-B.

SemEval-2018 Task 3: Irony Detection in English Tweets

“…For Task B, the top five is nearly similar to the top five for Task A and includes the following teams: UCDCC (Ghosh, 2018), NTUA-SLP (Baziotis et al, 2018), THU NGN (Wu et al, 2018), NLPRL-IITBHU (Rangwani et al, 2018) and NIHRIO (Vu et al, 2018). All of the teams tackled multiclass irony classification by applying (mostly) the same architecture as for Task A (see earlier).…”

“…ensemble LR models and concatenated word embeddings instead of averaged) model. NLPRL-IITBHU (Rangwani et al, 2018) ranked fourth and used an XGBoost Classifier to tackle Task A. They combined pre-trained CNN activations using DeepMoji (Felbo et al, 2017) with ten types of handcrafted features.…”

Detecting Hate Speech for Italian Language in Social Media

“…The motivation for this work comes from two directions. In a first place, the recent and promising results found by some authors Gonález et al, 2018;Rangwani et al, 2018;Peng et al, 2018) in the use of convolutional networks and recursive networks, also the hybridization of them for dealing with figurative language. The second direction is motivated by the wide use of linguistic features manually encoded which have showed to be good indicators for discriminating among ironic and non ironic content Reyes and Rosso, 2014;Barbieri et al, 2014;Farías et al, 2016;Farías et al, 2018).…”

Emory at WNUT-2020 Task 2: Combining Pretrained Deep Learning Models and Feature Enrichment for Informative Tweet Identification

“…Emoji Embedding: Emojis can succinctly represent emotional expressions and have become popular in social media. Recently, several studies have revealed that incorporating emoji information into deep learning models can benefit the performance of tweet classification tasks (Singh et al, 2019;Rangwani et al, 2018). Inspired by that, we employ a pretrained emoji encoder named emoji2vec 4 (Eisner et al, 2016) to convert emojis into emoji embeddings.…”