About this project, this project is a hands-on project which I had found on Coursera. I think this hands-on project is very reasonable project for those who already have a little knowledge of Neutral Network and want to learn deeper about Recurrent Neural Networks, Long Short-Term Memories. This project will give you an overview of the basic steps to build a DL model in general and a NLP model in particular; from basic steps such as: import library, load dataset, EDA, data cleaning, build and train model, access performance to specific steps of a NLP project like plot word-cloud, removing punctuation and stop words, tokenizing and padding on text corpus.
Me, after completed some DL courses on Coursera, I want to dig deeper more about NN, especially RNN and LSTM so that I choose this course. This course helps me consolidate my knowledge and gain new knowledge about natural language processing; understand the intuition behind RNN, LSTMs, how to prepare data by performing tokenization and padding.
The dataset include True.csv with the shape is (21417, 4) and Fake.csv with the shape is (23481, 4). Cause the size of this dataset is two large to upload on Github so I zipped it.
The detail process can be found in my notebook.
You can access full here: https://github.com/dchatca/Fake-news-classification/blob/main/fake-news-classification/fake_news_classification.ipynb
Bi-Directional RNN and LSTM with model accuracy : 0.9982