Skip to content

Latest commit

 

History

History
20 lines (12 loc) · 1.01 KB

README.md

File metadata and controls

20 lines (12 loc) · 1.01 KB
Raw

Topic-Detection-in-arabic

Bidirectional LSTM model with conv1D , Bidirectional LSTM model only and conv1D only , 3 models for testing the most efficient way for topic detection.

Bidirectional-LSTMs-model-with-conv1D

LSTMs neural networks has excelled in the process of sequence understanding. And combining it with conv1D has become the new state of art in the field of Topic Detection. I used keras implementation with tensorflow backend.

First-Section-in-the-Code

For people who will upload the dataset and ara2vec model on their google drive. It is a fastter way to work on google colab linked with the drive.

Steps-to-make-it-work.

1-You will find it step by step guidance in the Notebook in the rep.

2-You can download the dataset from the following link. https://drive.google.com/open?id=1n4XTmPzygP6AvO6ekST12G-zEDH9Wqyu

3-You can download the ara2vec model from the following link. https://github.com/bakrianoo/aravec

Any inquires about anything please email me on [email protected]

Happy learning.