LynyrdSkynyrd at WNUT-2020 Task 2: Semi-Supervised Learning for\n Identification of Informative COVID-19 English Tweets

2020/09/08 by Abhilasha Sancheti, Sancheti, Abhilasha, Kushal Chawla +3
Social Sciences · Computer Science · #Misinformation and Its Impacts #Hate Speech and Cyberbullying Detection #Sentiment Analysis and Opinion Mining

paper · pdf · doi:10.48550/arxiv.2009.03849

Abstract

We describe our system for WNUT-2020 shared task on the identification of\ninformative COVID-19 English tweets. Our system is an ensemble of various\nmachine learning methods, leveraging both traditional feature-based classifiers\nas well as recent advances in pre-trained language models that help in\ncapturing the syntactic, semantic, and contextual features from the tweets. We\nfurther employ pseudo-labelling to incorporate the unlabelled Twitter data\nreleased on the pandemic. Our best performing model achieves an F1-score of\n0.9179 on the provided validation set and 0.8805 on the blind test-set.\n

Related

• WNUT-2020 Task 2: Identification of Informative COVID-19 English Tweets
• CIA_NITT at WNUT-2020 Task 2: Classification of COVID-19 Tweets Using Pre-trained Language Models
• TATL at W-NUT 2020 Task 2: A Transformer-based Baseline System for Identification of Informative COVID-19 English Tweets
• EdinburghNLP at WNUT-2020 Task 2: Leveraging Transformers with\n Generalized Augmentation for Identifying Informativeness in COVID-19 Tweets
• NutCracker at WNUT-2020 Task 2: Robustly Identifying Informative\n COVID-19 Tweets using Ensembling and Adversarial Training