Online misinformation detection

Inspiration

This project is very much related to my field (Data Science). I am from computer science background with interest in Artificial Intelligence. This project was interesting as we can make a model which can learn to filter fake information or news and it can become more and more accurate as we feed the data.

What it does

My model will help find out if the given information is either real or fake with accuracy range between (0-1). 0 being the least(misinformation) and 1 being the highest(genuine information). It accepts the input paragraph and find out with respect to previous information and outputs the result.

How we built it

We will be using LSTM(Long short term memory) which is an artificial recurrent neural network. It is capable of learning order dependence in sequence prediction. There are two datasets from Kaggle consisting the real and fake data which we will using to train our model. The real dataset consists of the sources or tweets or Reuters so that we can distinguish between the real and fake for a model. The model will be able to predict the accuracy of the given information.

Challenges we ran into

Not enough data to train the model to a greater accuracy and some information is being classified wrong such as real information with lower accuracy.

Accomplishments that we're proud of

I am proud that my model is able to predict and successfully completed without errors but need to improve its accuracy.

What we learned

I have learned a lot such as understanding the data before in order to build a model. Visualizing data and understanding it would be the best thing before you go forward.

What's next for Online misinformation detection

I would like to improve the model's accuracy with more data and deploy it on cloud(create an application).

Built With

• ai
• google-colab
• machine-learning
• python