https://github.com/taivop/joke-dataset
In this project I will atempt to:
- Do exploratory data analysis.
- Create a word-level NLP data preprocessing pipeline.
- Build a (bidirectional) recurrent deep learning model.
- Build a 1d convolutional deep learning model.
- Build a hybrid (recurrent + convolutional) deep learning model.
- Compare the performance of the different deep learning models.
- Implement plaing model without Glove embeddings
- Try data balancing on train dataset
@misc{pungas,
title={A dataset of English plaintext jokes.},
url={https://github.com/taivop/joke-dataset},
author={Pungas, Taivo},
year={2017},
publisher = {GitHub},
journal = {GitHub repository}
}