Deep Learning and Word Embeddings for Tweet Classification for Crisis Response
Abstract
Tradition tweet classification models for crisis response focus on convolutional layers and domain-specific word embeddings. In this paper, we study the application of different neural networks with general-purpose and domain-specific word embeddings to investigate their ability to improve the performance of tweet classification models. We evaluate four tweet classification models on CrisisNLP dataset and obtain comparable results which indicates that general-purpose word embedding such as GloVe can be used instead of domain-specific word embedding especially with Bi-LSTM where results reported the highest performance of 62.04% F1 score.
Cite
@article{arxiv.1903.11024,
title = {Deep Learning and Word Embeddings for Tweet Classification for Crisis Response},
author = {Reem ALRashdi and Simon O'Keefe},
journal= {arXiv preprint arXiv:1903.11024},
year = {2019}
}
Comments
This paper has been accepted and presented in the 3rd National Computing Colleges Conference (NC3) in Abha, Saudi Arabia on 9th October 2018