The COOL Kids

Inspiration

Fake news spreads misinformation that erodes public trust, fuels fear and division, and disrupts democratic processes. It misleads people, damages reputations, and threatens societal harmony. Combating it is essential for a well-informed and united society. With misinformation spreading faster than ever, we felt the need to create a tool that helps users distinguish between real and fake news. Our goal was to build a smart, accessible, and reliable system that supports informed decision-making in a digital world overloaded with content.

What it does

Our software uses advanced machine learning techniques to detect fake news. Given a news article or headline, it analyzes the text and classifies it as either real or fake. The system leverages a combination of text processing (TF-IDF) and a custom optimization algorithm (Logistic Regression & Orthogonal Atom Optimization Strategy - LROAOS) for highly accurate classification.

How we built it

Dataset: We used labeled datasets from real and fake news articles.
Text Processing: We applied TF-IDF to convert textual data into numerical features.
Model: We implemented logistic regression, optimized using OAOS to improve feature weighting.
Enhancement: We integrated orthogonal learning strategies and a custom energy-based search mechanism for optimization.
GUI: We used python Tkinter to provide user-friendly and attractive graphic

Challenges we ran into

Not getting enough sleep

Accomplishments that we're proud of

Successfully integrating a novel optimization algorithm (OAOS) into a classification model.
Achieving competitive accuracy on real-world data.
Building a fully functional pipeline from raw text to prediction.

What we learned

How optimization techniques like OAOS can enhance traditional models.
Collaborating efficiently as a team and overcoming workloads.

What's next for The COOL Kids

Expanding the dataset to include multilingual and more nuanced examples. Moreover, to include Incorporating explainable AI features to show users why a piece of news is flagged.