This study aims to explore the application of Artificial Neural Networks (ANN) in spatial analysis of Geographic Information Systems (GIS) to improve the accuracy and efficiency of spatial data analysis. By combining ANN with GIS, the analysis focuses on land use and environmental monitoring data, especially the prediction of air quality index (AQI). Traditional GIS methods have limitations in dealing with complex nonlinear and high-dimensional data, while ANN can effectively solve these problems through its self-learning and self-adaptive capabilities. The study performed detailed pre-processing of raw data, including data cleaning, standardisation and normalisation to ensure data quality and consistency. These data, including land use type, surface temperature, AQI and population density, were converted into numerical vectors suitable for ANN model processing, laying the foundation for model training. For model construction, the BP neural network model was chosen, and the network structure and parameters were optimised through experiments. The number of nodes in the input layer is consistent with the feature dimension, the number of nodes in the hidden layer is determined by experiment, and the number of nodes in the output layer is set according to the task requirements. During the training process, a back-propagation algorithm is used to continuously adjust the weights and biases to minimise the mean square error (MSE). To prevent overfitting, the study introduces L2 regularisation and cross-validation methods to improve the generalisation ability of the model. After the training was completed, the predictive performance of the model was evaluated using the test set data, and the results showed that the model exhibited high accuracy in the AQI prediction task, and the predicted values were highly consistent with the actual values, which verified the effectiveness of ANN in GIS spatial analysis. The study also mapped the prediction results onto geospatial space through spatial visualisation techniques to generate AQI distribution maps for each region, which visually demonstrated the spatial distribution of air quality and provided an important reference for environmental management and decision-making. The AQIs of industrial areas and high population density areas are higher, while those of woodlands and waters are lower, which is consistent with the actual geographic features and human activity patterns. Despite the remarkable results of the study, there are still some limitations. Data quality and data quantity have a direct impact on model performance, and future research should consider collecting more diverse and larger data sets. In addition, the ‘black box’ nature of the ANN model makes it difficult to explain its decision-making process, and future research should explore methods that incorporate explanatory techniques to improve the transparency and interpretability of the model.

Xingxin Yang, Yongkang Wang, Ke Lu, Youquan Wu, Di Zhao. Artificial neural network modelling in GIS spatial analysis. Academic Journal of Computing & Information Science (2024), Vol. 7, Issue 6: 32-37. https://doi.org/10.25236/AJCIS.2024.070605.

[1] Mollalo, A., Mao, L., Rashidi, P., & Glass, G. E. (2019). A GIS-based artificial neural network model for spatial distribution of tuberculosis across the continental United States. International journal of environmental research and public health, 16(1), 157.

[2] Lee, S., Ryu, J. H., Min, K., & Won, J. S. (2003). Landslide susceptibility analysis using GIS and artificial neural network. Earth Surface Processes and Landforms: The Journal of the British Geomorphological Research Group, 28(12), 1361-1376.

[3] Biswajeet, P., & Saro, L. (2007). Utilization of optical remote sensing data and GIS tools for regional landslide hazard analysis using an artificial neural network model. Earth Science Frontiers, 14(6), 143-151.

[4] Noack, S., Knobloch, A., Etzold, S. H., Barth, A., & Kallmeier, E. (2014). Spatial predictive mapping using artificial neural networks. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 40, 79-86.

[5] Lee, S., Song, K. Y., Kim, Y., & Park, I. (2012). Regional groundwater productivity potential mapping using a geographic information system (GIS) based artificial neural network model. Hydrogeology Journal, 20(8), 1511.

[6] Maithani, S., Jain, R. K., & Arora, M. K. (2007). An artificial neural network based approach for modelling urban spatial growth. ITPI Journal, 4(2), 43-51.

[7] Pradhan, B., & Lee, S. (2009). Landslide risk analysis using artificial neural network model focusing on different training sites. Int J Phys Sci, 3(11), 1-15.

[8] Pradhan, B., Lee, S., & Buchroithner, M. F. (2010). A GIS-based back-propagation neural network model and its cross-application and validation for landslide susceptibility analyses. Computers, Environment and Urban Systems, 34(3), 216-235.

[9] Khoirunisa, N., Ku, C. Y., & Liu, C. Y. (2021). A GIS-based artificial neural network model for flood susceptibility assessment. International Journal of Environmental Research and Public Health, 18(3), 1072.

[10] Mokarram, M., Taripanah, F., & Pham, T. M. (2024). Spatial-temporal analysis of atmospheric environment in urban areas using remote sensing and neural networks. Sustainable Computing: Informatics and Systems, 42, 100987.

[11] Krivoguz, D. (2024). Geo-spatial analysis of urbanization and environmental changes with deep neural networks: Insights from a three-decade study in Kerch peninsula. Ecological Informatics, 102513.