ForeXcast

I devised a program for predicting exchange rates of any given currency using deep neural networks combined with time-series forecasting models. This involved extensive data preprocessing and feature engineering to ensure the model could accurately interpret historical currency trends. By leveraging the Tradermade API, I fetched historical currency price data, which was essential for training the model. I then implemented and fine-tuned a deep neural network, optimizing it to capture complex patterns and dependencies in the time-series data.

The result was a highly accurate prediction model, achieving a 97.5% accuracy rate in forecasting future currency prices. This program provides users with valuable insights and predictions, aiding in making informed financial decisions. Additionally, I incorporated a user-friendly interface to facilitate easy access to predictions, ensuring that users can seamlessly utilize the tool for their currency trading or financial planning needs.