Using Deep Learning to Predict Global Population Dynamics and Construction Risks

Authors

  • Kai Jin
  • Xingxin Yang
  • Yonghao Zhou
  • Yongzheng Wu

DOI:

https://doi.org/10.62051/3my9gm68

Keywords:

Long Short-Term Memory Networks (LSTM); Global Population Projections; Country Risk Ratings; Time Series Analysis; Error Autocorrelation.

Abstract

In this study, a long short-term memory network (LSTM) model is used to predict the population size and risk level of each country in 2030 by combining global population data from 1950 to 2021 and country risk rating indices from 1954 to 2023.The core of the LSTM model is its three gating units: forgetting gates, input gates, and output gates, which enable the model to effectively deal with the long-term dependence problems and avoid the problems of gradient vanishing and gradient explosion in traditional recurrent neural networks. Through correlation analysis of the input data and prediction errors, as well as an autocorrelation study of the errors, the model shows prediction accuracy at certain time delays, although the prediction errors of the model are not observed to be completely random on certain delay terms, implying that the model may need further optimisation in these areas. In addition, this study applies the entropy method to calculate the construction suitability evaluation index for each country in 2030, which provides a scientific basis for the development of future construction strategies. By analysing the mean square error (MSE) during the training process, the study shows that the performance of the model on the training and validation sets gradually improves as the number of training rounds increases, although the MSE on the test set increases at some points, showing signs of overfitting. Finally, countries were scored and ranked for construction suitability based on predicted population and risk class scores. Based on the composite suitability score index, we categorised the countries into three building suitability classes, each corresponding to a different building strategy. In areas of low suitability, robust materials and designs that are resistant to natural hazards are recommended; in areas of moderate suitability, sustainable building design and green building materials are recommended; and in areas of high suitability, investment in high-quality and high-end design building projects should be emphasised to meet the needs of higher-income groups. These strategies will contribute to the needs of different regions and populations and improve the accuracy and efficiency of real estate decisions. Through the application of this deep learning model, we have not only improved our ability to predict global population dynamics and construction risks, but also provided important data support for urban planning and development strategies on a global scale.

Downloads

Download data is not yet available.

References

Lin, S. S., Shen, S. L., Zhou, A., & Xu, Y. S. (2021). Risk assessment and management of excavation system based on fuzzy set theory and machine learning methods. Automation in Construction, 122, 103490.

Yakovlev, S., Bazilevych, K., Chumachenko, D., Chumachenko, T., Hulianytskyi, L., Meniailov, I., & Tkachenko, A. (2020). The concept of developing a decision support system for the epidemic morbidity control.

Pourhomayoun, M., & Shakibi, M. (2021). Predicting mortality risk in patients with COVID-19 using machine learning to help medical decision-making. Smart health, 20, 100178.

Zhang, L., Wang, Y., Niu, M., Wang, C., & Wang, Z. (2020). Machine learning for characterizing risk of type 2 diabetes mellitus in a rural Chinese population: The Henan Rural Cohort Study. Scientific reports, 10(1), 4406.

Berrang-Ford, L., Sietsma, A. J., Callaghan, M., Minx, J. C., Scheelbeek, P. F., Haddaway, N. R., ... & Dangour, A. D. (2021). Systematic mapping of global research on climate and health: a machine learning review. The Lancet Planetary Health, 5(8), e514-e525.

Quiroz-Juárez, M. A., Torres-Gómez, A., Hoyo-Ulloa, I., León-Montiel, R. D. J., & U’Ren, A. B. (2021). Identification of high-risk COVID-19 patients using machine learning. Plos one, 16(9), e0257234.

Zhao, H., Zhang, X., Xu, Y., Gao, L., Ma, Z., Sun, Y., & Wang, W. (2021). Predicting the risk of hypertension based on several easy-to-collect risk factors: a machine learning method. Frontiers in Public Health, 9, 619429.

Pang, L., Wang, K., Tao, Y., Zhi, Q., Zhang, J., & Lin, H. (2021). A new model for caries risk prediction in teenagers using a machine learning algorithm based on environmental and genetic factors. Frontiers in Genetics, 12, 636867.

Yang, J., Ju, J., Guo, L., Ji, B., Shi, S., Yang, Z., ... & Yuan, P. (2022). Prediction of HER2-positive breast cancer recurrence and metastasis risk from histopathological images and clinical information via multimodal deep learning. Computational and structural biotechnology journal, 20, 333-342.

Pham, B. T., Luu, C., Van Dao, D., Van Phong, T., Nguyen, H. D., Van Le, H., ... & Prakash, I. (2021). Flood risk assessment using deep learning integrated with multi-criteria decision analysis. Knowledge-based systems, 219, 106899.

Downloads

Published

20-06-2024

How to Cite

“Using Deep Learning to Predict Global Population Dynamics and Construction Risks” (2024) Transactions on Computer Science and Intelligent Systems Research, 4, pp. 22–29. doi:10.62051/3my9gm68.

Similar Articles

1-10 of 64

You may also start an advanced similarity search for this article.