This project demonstrates the real-world application of machine learning in social media analytics. I designed and developed an interactive web application that predicts the future follower and like counts of Instagram users while also categorizing them into influencer tiers using clustering techniques.
The platform leverages Random Forest Regression to forecast growth metrics like followers and likes based on key inputs such as average likes per post, number of posts, and engagement rate. Additionally, KMeans Clustering is used to automatically classify influencers into one of three data-driven categories: Micro Influencer, Mid-tier, or Celebrity.
The frontend was built using Streamlit, offering a clean and intuitive interface where users can enter their current statistics and receive instant predictions, influencer categorization, and personalized recommendations. Visualizations such as bar charts and pie charts help users understand their performance metrics and potential for growth. Logical caps were included to ensure realistic and grounded predictions.
This project showcases my ability to integrate machine learning models into full-stack solutions and deliver meaningful, user-friendly applications. It reflects skills in predictive analytics, data visualization, model deployment, and building intelligent interfaces that bridge data science and real-world impact.
← Back to Portfolio