Multi-Task Learning for Pun Detection and Location with BERT and Graph Convolutional Neural Networks
BERT_GNN_MTL_Control_BiLSTM_v4.ipynb
https://alt.qcri.org/semeval2017/task7/
This is the Final Project of Group 4, COSC-572 Empirical Methods in Natural Language Processing, Spring 2023, Georgetown University
Course Webpage:
https://people.cs.georgetown.edu/nschneid/cosc572/s23/index.html
BERT_GNN_MTL_Control_BiLSTM_v1.ipynb
For this subtask, participants are given an entire raw data set. For each context, the system must decide whether or not it contains a pun.
For this subtask, the contexts not containing puns are removed from the data set. For each context, the system must identify which word is the pun.
Google Colab (with Python 3)
https://colab.research.google.com/
The file name of each file contains the models which it used.
All the files can be run in Google Colab.
For the Evaluation.ipynb, you can load a model (.pt file) from Google Drive and do the evaluation.
https://alt.qcri.org/semeval2017/task7/data/uploads/semeval2017_task7.tar.xz
From Yang et al. (2015)
(This dataset is not publicly available and was obtained by contacting paper authors directly.)