Nowadays, the majority of the world's population has an access to the information, internet is one of the easiest way to share massive information no matter if this information is certified or not. In this Covid-19 period, The medias have a crucial role concerning prevention and disease control. They need to provide certified and accurate information.
What it does
Our project focus on a way to enable people to have access to certified informations.
We developed a twitter chatbot which will be able to retweet certified informations about Covid-19.
You have also the possibility to ask daily information about a specific country. It's really easy :
- Write a tweet and tag our chatbot @Covidbot5 if you want to look at the evolution of Covid-19 in a specific country.
How I built it
The chatbot was built with tweepy. This chatbot is connected to a BLSTM neural network designed with Keras. This neural network is able to classify fake and certified informations.
Challenges I ran into
We had to manage a large dataset and fill this dataset with Covid-19 certified data.
Accomplishments that I'm proud of
We are proud to give to people an easy access to certified news thanks to this twitter chatbot.
What I learned
During this challenge we had the opportunity to develop a chatbot for the very first time. It was also our first time with NLP.
What's next for Fighting Fake News related to COVID-19 using Deep Learning
We would like to improve the classification power of our chatbot thanks to new databases or new Deep Learning models.