Emotional Tweets: A Convolutional-BiLSTM Approach to Emotion Classification

DOI

10.2991/978-94-6463-262-0_74How to use a DOI?

Keywords

emotion classification; Machine learning model; neural network; BiLSTM; social media analysis; Twitter data; text data processing

Abstract

We introduce a neural network model using a convolutional and bidirectional long short-term memory framework for the purpose of emotion classification. Our work utilizes a corpus of more than 200,000 English tweets sourced from the Twitter API, collected between January 2020 and September 2021, related to the COVID-19 pandemic. By employing pre-trained 50-dimensional GloVe word embeddings for vectorizing the short-form text data, our model achieves a significantly higher accuracy than what would be anticipated from random classification. It demonstrates a 65.39% success rate in categorizing each tweet into one of the five key emotions, markedly outperforming the random baseline of 41.5%. Along with discussing pertinent previous studies that have shaped our model's design, we also detail the necessary steps for data acquisition and processing to generate the data used, as well as the method for developing, training, and fine-tuning the model.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Hongyuan Ran
PY  - 2023
DA  - 2023/10/09
TI  - Emotional Tweets: A Convolutional-BiLSTM Approach to Emotion Classification
BT  - Proceedings of the 3rd International Conference on Management Science and Software Engineering (ICMSSE 2023)
PB  - Atlantis Press
SP  - 708
EP  - 720
SN  - 2589-4943
UR  - https://doi.org/10.2991/978-94-6463-262-0_74
DO  - 10.2991/978-94-6463-262-0_74
ID  - Ran2023
ER  -