NFT Sentiment Price Predictor

Overview

The NFT Sentiment Price Predictor is an innovative platform that combines real-time market data with social media sentiment analysis to predict NFT price movements. We implemented and optimized over 20 functional components for NFT price charts overlay with predicted prices and sentiment over customizable time series parameters.

This project leverages advanced natural language processing techniques to analyze social media discussions, news articles, and community forums to gauge market sentiment around specific NFT collections. By combining this sentiment data with historical price information and market trends, the platform provides users with predictive insights to make more informed investment decisions in the volatile NFT market.

Challenges

• •Processing and analyzing large volumes of unstructured social media data in real-time
• •Developing accurate sentiment analysis models for the specialized language used in NFT and crypto communities
• •Creating intuitive visualizations for complex data relationships between sentiment and price movements
• •Building a responsive and performant frontend that can handle real-time data updates
• •Integrating with multiple blockchain APIs to gather comprehensive NFT market data

Solutions

• •Implemented a distributed data processing pipeline using PySpark to handle large volumes of social media data
• •Developed custom NLP models using SparkNLP specifically trained on NFT and crypto terminology
• •Created interactive chart components with customizable parameters using TypeScript and Next.js
• •Built a responsive frontend with Tailwind CSS that adapts to different screen sizes and devices
• •Developed a unified API layer using tRPC to integrate data from multiple blockchain sources

Results

• •Successfully completed NUS Orbital Apollo requirements.

Gallery