Deep learning for sarcasm detection on Twitter

Ostos Balfagón, Rubèn

Deep learning for sarcasm detection on Twitter

License

Citation

Ostos Balfagón, R. Deep learning for sarcasm detection on Twitter. 2020. handle: http://hdl.handle.net/10230/45891

This citation was generated automatically.

Abstract

Automatic sarcasm detection has come very helpful to improve sentiment analysis, being proved their direct relationship recently. Numerous deep learning approaches have been applied lately in this task obtaining very interesting results. It is common to find sarcastic messages in social networks, Twitter for example has a good API that comes handy for natural language processing tasks. From the point of view of companies and organizations it is interesting to improve their sentiment analysis models in order to have a better understanding of possible clients or even society. The objective of this project is to analyze the most notorious work in this task up to the moment and test some deep learning architectures with different word embeddings, giving a lead on how to detect sarcasm on Twitter with good results using the Python library TensorFlow.