Skip to content

Latest commit

 

History

History
20 lines (11 loc) · 425 Bytes

README.md

File metadata and controls

20 lines (11 loc) · 425 Bytes

Arabic Sentiment Analysis

This project was for the neural networks and deep learning course

Preprocessing 1- Remove duplicates

2- Remove punctuatuin.

3- Remove any numbers and any word that contains a number.

4- Remove any word that is not Arabic.

5- Remove the emoji and put a word that explains this emoji.

6- Remove stop words.

7- Apply tokenization to the text.

Apply the LSTM and Transformer model.