JU_ETCE_17_21 at SemEval-2019 Task 6: Efficient Machine Learning and Neural Network Approaches for Identifying and Categorizing Offensive Language in Tweets

Abstract

This paper describes our system submissions as part of our participation (team name: JU ETCE 17 21) in the SemEval 2019 shared task 6: "OffensEval: Identifying and Categorizing Offensive Language in Social Media". We participated in all the three sub-tasks: i) Sub-task A: offensive language identification, ii) Sub-task B: automatic categorization of offense types, and iii) Sub-task C: offense target identification. We employed machine learning as well as deep learning approaches for the sub-tasks. We employed Convolutional Neural Network (CNN) and Recursive Neural Network (RNN) Long Short-Term Memory (LSTM) with pre-trained word embeddings. We used both word2vec and Glove pre-trained word embeddings. We obtained the best F1score using CNN based model for sub-task A, LSTM based model for sub-task B and Logistic Regression based model for sub-task C. Our best submissions achieved 0.7844, 0.5459 and 0.48 F1-scores for sub-task A, sub-task B and sub-task C respectively.

Classification of Imbalanced Offensive Dataset – Sentence Generation for Minority Class with LSTM

“…In this RNN architecture, uni-gram and bi-gram features were used and in a fully connected (FC) layer, seven different machine learning models were performed. Pal et al applied different machine learning and deep learning algorithms [37]. While the best among machine learning algorithms was LR-tri-gram with 0.7231 F-score, among deep learning algorithms the best was CNN-Glove with 0.7844 F-score.…”