Inspiration
I love to work with machine learning, especially Natural Language Processing. Fake news is a big problem in the digital age, and even younger people are being exposed to suspicious information on the internet. Thus, I decided to create this to prevent people from trusting fake news. For creativity, I decided to name the project after Master Shifu from Kung Fu Panda, as his name rhymed well with the "Sham News" I was looking at.
What it does
The website takes in the author's name and the article's title. Then, using a machine learning model I created, the website predicts whether or not the article is likely to be fake news.
How we built it
I used a dataset from Kaggle for training and testing datasets. I then took the author's name and the title of the article for each news article, removed stopwords (words that provide no real information to the news article), stemmed the words (remove prefixes and suffixes, reducing the words down to their roots), and then converted the words into vectors (numerical data). I then ran a logistic regression machine learning model on the data and created a Flask website (also using HTML, CSS, and JS) to put the machine learning model to work.
Challenges we ran into
Once I successfully created the machine learning model, I ran into some trouble forming the actual website because I had never really worked with frontend before.
Accomplishments that we're proud of
I am really proud to have finished this product on my own!
What we learned
I learned how to use Flask!
What's next for Shifu Sham News
I tried to actually use the text of the article in training and testing the machine learning model, but the program simply took too long to run; there did not seem to be any programming errors though. In the future, I would like to properly implement the text into the machine learning model, yet still have a relatively short runtime.
Log in or sign up for Devpost to join the conversation.