Sarcasm detection model

Sarcasm detection model using a 1D CNN with an embedding layer, a global average pooling layer, and two dense layers. The model was trained on the News Headlines Dataset for Sarcasm Detection and was able to achieve an accuracy of 83% on the testing dataset. It also achieved an accuracy of 88% on the K-Fold cross-validation dataset.

Links to the code:

• Github
• Kaggle