کنفرانس مهندسی برق ایران

صفحه اصلی / سی و سومین کنفرانس بین المللی مهندسی برق

Crypto Currency Price Prediction Using Preprocessed Scaled Inputs LSTM Model Enhanced by Improved Gray Wolf Optimization

نویسندگان :

Amir RabbaniParsa¹ Mahboobeh Hoshmand² Seyyed Abed Hosseini³

1- دانشگاه آزاد واحد مشهد 2- دانشگاه آزاد واحد مشهد 3- دانشگاه آزاد واحد مشهد

کلمات کلیدی :

Bitcoin،Long Short-Term Memory،Gray Wolf Optimization،Prediction،Hyperparameter،Forecasting

چکیده :

This study introduces a novel approach for forecasting cryptocurrency prices by integrating Long Short-Term Memory (LSTM) networks with an Improved Gray Wolf Optimization (iGWO) algorithm. The hybrid model enhances predictive accuracy by leveraging the LSTM model's proficiency in capturing time-series dependencies and the iGWO's effectiveness in hyperparameter optimization. Experimental evaluations using metrics such as Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) demonstrate the superior performance of the proposed model over traditional forecasting methods. The findings offer valuable insights for financial analysts and investors aiming to mitigate risks associated with the volatile cryptocurrency market.