Financial Predictor using Sentiment Analysis via OpenAI API (for Maxx Source)

Designing an LLM-Powered Market Sentiment Engine for Stock & Crypto Trend Prediction

PROBLEM

STATEMENT

The client needed to perform sentiment analysis on financial news and social media content (posts, videos, etc.). They wanted to gather as much data as possible on their specified keywords to provide useful insights about the market trends and rates of certain stocks and cryptocurrencies. They were looking for a data engineer with expertise in LLM and OpenAI API to help with this project.

SOLUTION

I developed a multi-platform data pipeline which gathers relevant information from YouTube, Twitter (X), and Google Trends and stores it in PostgreSQL. The data was then cleaned and fed to different models of GPT to analyze which of them gives the most suitable results. Based on the results, we decided to move ahead with GPT-4o.

The efficiency of the model continued to improve as more and more data was used for training. The end product was able to provide a lot of value as several of their clients made profitable decisions in stock and crypto market.

Input

Keywords for extracting data

Output

A fully-trained sentiment analysis system for predicting the trajectory of businesses, stocks, and cryptocurrencies