Fake Review Checker

• 

  Text Input functionality

• 

  File Upload functionality

• 

  Word Cloud functionality

Inspiration

I have noticed over the past few years that there has been a lot of fake reviews on the internet and they are getting increasingly sophisticated and difficult to detect. So, I have decided to built a review checked to detect such fake reviews.

What it does

This web application classifies if the review(s) provided via text or file in CSV or JSON format are real or fake. A confidence rating is provided upon classification, determining how confident the mode is in classifying the reviews.

How we built it

The web application was built using Python and Streamlit to host the application. The model was trained and built using a fake review dataset and scikit-learn using a Logistic Regression model.

Challenges we ran into

The challenge is mainly building and fine-tuning the model as well as preprocessing the data to train the model.

Accomplishments that we're proud of

I am proud to have build an functional model to check reviews in a practical way.

What we learned

This is my first experience training and building a model for a practical application.

What's next for Fake Review Checker

I will continue to improve on the model to increase the confidence ratings as well as adding in new features such as bulk file upload and showing more visual summaries.

Built With

• juypter
• python
• scikit-learn
• streamlit