Predicting population growth in Libya using deep learning techniques (LSTM)

Mansour, Walid W Ramadan and Alkharif, Ali Ramadan Mustafa and Atomi, Abdulrazag Mukhtar Elmahdi (2025) Predicting population growth in Libya using deep learning techniques (LSTM). World Journal of Advanced Research and Reviews, 25 (2). pp. 291-295. ISSN 2581-9615

Predicting population growth is a crucial element in planning future resources and sustainable development. With rapid changes in global demographics, there is a growing need for robust and accurate forecasting models. This research introduces a model leveraging deep learning techniques to analyze historical and demographic data for predicting population growth. Specifically, the study implements a Long Short-Term Memory (LSTM) neural network to address the temporal dynamics of population data. Performance enhancements were achieved through advanced techniques, such as feature embedding and algorithmic optimization. The study also explores the challenges of population prediction in regions with fluctuating growth patterns and demonstrates the model’s ability to outperform traditional methods like ARIMA. Results show that the proposed model achieves high accuracy, providing valuable insights for policymakers and planners. The integration of deep learning approaches highlights their potential to revolutionize population growth forecasting and support strategic decision-making.