Related Experiment Videos
House price prediction using a hybrid GRU-MLP based on binary whale optimization algorithm and ant colony
Sarah M Alhammad1, Yasser Fouad2, Amira A Mahmoud3
1Department of Computer Sciences, College of Computer and Information Sciences, Princess Nourah Bint Abdulrahman University, 11671, Riyadh, Saudi Arabia.
Scientific Reports
|July 25, 2026
Summary
This study introduces an optimized deep learning model for accurate house price prediction. The novel framework enhances real estate valuation by improving prediction accuracy and model stability.
Area of Science:
- Artificial Intelligence
- Machine Learning
- Real Estate Analytics
Background:
- Accurate house price prediction is crucial for real estate valuation and investment.
- Existing models often lack the precision needed for complex property attributes and market dynamics.
Purpose of the Study:
- To develop an optimized hybrid deep learning framework for enhanced house price prediction.
- To integrate advanced optimization algorithms for feature selection and hyperparameter tuning.
Main Methods:
- A hybrid Gated Recurrent Unit (GRU) and Multilayer Perceptron (MLP) model was developed.
- Binary Whale Optimization Algorithm (BWOA) was used for feature selection.
- Ant Colony Optimization (ACO) was employed for hyperparameter tuning.
- A Kaggle house price dataset with 500 records was utilized for evaluation.
Main Results:
- The proposed BWOA-ACO-GRU-MLP model significantly outperformed standalone GRU, MLP, CNN, LSTM, and BiLSTM models.
- Achieved high accuracy with an R-squared of 99.04% and low error metrics (MSE: 0.0146, MAE: 0.1051).
Conclusions:
- The hybrid deep learning framework with integrated optimization techniques offers superior performance for house price estimation.
- This data-driven approach provides a reliable tool for smart real estate valuation applications.