Forecasting and Spatial Analysis of Sustainable Development Goals 1 and 4 Using TabNet
DOI:
https://doi.org/10.59395/ijadis.v7i2.1535Keywords:
TabNet , Sustainable Development Goals , Choropleth Map , Coefficient of Determination, Predictive ModelingAbstract
This study applies the Tabular Neural Network (TabNet) to forecast and spatially analyze Sustainable Development Goals (SDGs) 1 and 4, providing evidence-based support for policymaking aligned with the United Nations 2030 Agenda. The dataset includes 3,841 observations from 167 countries, combining World Bank socio-economic indicators as predictors with SDG index scores as targets. A country-based split allocated 80% of countries for training and 20% for testing, ensuring cross-country generalization. Future projections for 2025–2030 use a persistence-based scenario in which socio-economic covariates remain at their most recent values, and lagged SDG scores are recursively updated, producing conditional simulations rather than purely autoregressive forecasts. The TabNet model demonstrates strong predictive performance: SDG 1 achieves RMSE of 4.22 (train) and 3.99 (test) with R² of 0.89 and 0.92, while SDG 4 achieves RMSE of 3.89 and 3.76 with R² of 0.95 and 0.93. Choropleth mapping highlights regional disparities, with higher predicted scores in Europe and North America compared to Sub-Saharan Africa. These results indicate that TabNet, combined with spatial visualization, provides an interpretable, robust, and policy-relevant framework for monitoring global SDG progress.
Downloads
References
[1] S. Ark and T. Pfister, TabNet: Attentive Interpretable Tabular Learning, 35th AAAI Conf. Artif. Intell. AAAI 2021, vol. 8A, pp. 66796687, 2021, doi: 10.1609/aaai.v35i8.16826 DOI: https://doi.org/10.1609/aaai.v35i8.16826
[2] S. Joddy, Comparative Analysis of CNN, LSTM, and CNNLSTM for Indonesian Stock Prediction, Eng. Math. Comput. Sci. J., vol. 7, no. 3, pp. 283289, 2025, doi: 10.21512/emacsjournal.v7i3.14326 DOI: https://doi.org/10.21512/emacsjournal.v7i3.14326
[3] T. Liu, J. Fan, G. Li, N. Tang, and X. Du, Tabular Data Synthesis with Generative Adversarial Networks: Design Space and Optimizations, VLDB J., vol. 33, no. 2, pp. 255280, 2024, doi: 10.1007/s00778-023-00807-y DOI: https://doi.org/10.1007/s00778-023-00807-y
[4] United Nations Department of Economic and Social Affairs, Global Sustainable Development Goals Report 2023, 2023.
[5] M. de Carvalho Marques, A. A. Mohamed, and P. Feitosa, Sustainable Development Goal 6 Monitoring through Statistical Machine Learning Random Forest Method, Clean. Prod. Lett., vol. 8, no. December, p. 100088, 2025, doi: 10.1016/j.clpl.2024.100088 DOI: https://doi.org/10.1016/j.clpl.2024.100088
[6] S. U. Parvathy, V. K. Kolil, R. Raman, R. Vinuesa, and K. Achuthan, Integrating Sustainable Development Goals into Life Cycle Thinking: A Multidimensional Approach for Advancing Sustainability, Environ. Dev. Sustain., pp. 139, 2025, doi: 10.1007/s10668-024-05810-z DOI: https://doi.org/10.1007/s10668-024-05810-z
[7] A. Osman et al., Spatial Analysis of Synergies and Trade-Offs between the Sustainable Development Goals (SDGs) in Africa, Geogr. Sustain., vol. 3, no. 3, pp. 220231, 2022, doi: 10.1016/j.geosus.2022.07.003 DOI: https://doi.org/10.1016/j.geosus.2022.07.003
[8] K. Chenary and O. P. Kalat, Forecasting Sustainable Development Goals Scores by 2030 Using Machine Learning Models, Wiley Encycl. Manag., no. April, pp. 65206538, 2024, doi: 10.1002/sd.3037 DOI: https://doi.org/10.1002/sd.3037
[9] P. Utama, A. P. Wibawa, A. N. Handayani, and M. Y. Chuttur, Exploring the Role of Deep Learning in Forecasting for Sustainable Development Goals: A Systematic Literature Review, Int. J. Robot. Control Syst., vol. 4, no. 1, pp. 365400, 2024, doi: 10.31763/ijrcs.v4i1.1328 DOI: https://doi.org/10.31763/ijrcs.v4i1.1328
[10] A. Garc, R. A. Barrio, C. Gershenson, and K. K. Kaski, Sustainable Visions: Unsupervised Machine Learning Insights on Global Development Goals, arXiv preprint, pp. 123, 2025.
[11] M. S. U. Miah et al., Predicting Short-Term Energy Demand in the Smart Grid: A Deep Learning Approach for Integrating Renewable Energy Sources in Line with SDGs 7, 9, and 13, arXiv preprint, pp. 138, 2024. DOI: https://doi.org/10.2139/ssrn.4534111
[12] A. Esan et al., Long-Short-Term Memory Model for Fake News Detection in Nigeria, Ianna J. Interdiscip. Stud., vol. 5, no. 1, pp. 167180, 2023.
[13] N. Yang, Y. Wang, F. Wu, and Z. Wei, MapColorAI: Designing Contextually Relevant Choropleth Map Color Schemes Using a Large Language Model, Cartogr. Geogr. Inf. Sci., pp. 131, 2025. DOI: https://doi.org/10.1080/15230406.2025.2531055
[14] J. Reades and S. J. Rey, Geographical Python Teaching Resources: geopyter, J. Geogr. Syst., pp. 579597, 2021, doi: 10.1007/s10109-021-00346-6 DOI: https://doi.org/10.1007/s10109-021-00346-6
[15] J. Ha, M. Kambe, and J. Pei, Data Mining: Concepts and Techniques, 3rd ed. San Francisco, CA, USA: Morgan Kaufmann, 2011. doi: 10.1016/C2009-0-61819-5 DOI: https://doi.org/10.1016/C2009-0-61819-5
[16] T. Shi and K. Shide, A comparative analysis of LSTM, GRU, and Transformer models for construction cost prediction with multidimensional feature integration, J. Asian Archit. Build. Eng., 2025, doi: 10.1080/13467581.2025.2455034 DOI: https://doi.org/10.1080/13467581.2025.2455034
[17] N. Saka and S. M. Krishna, Efficient Feature Extraction Based on Optimized Gated Recurrent Unit Recurrent Neural Network for Automatic Thyroid Prediction, 2022, Research Square Platform LLC. doi: 10.21203/rs.3.rs-2319272/v1 DOI: https://doi.org/10.21203/rs.3.rs-2319272/v1
[18] L. L. Moreira, M. M. de Brito, and M. Kobiyama, Effects of different normalization, aggregation, and classification methods on the construction of flood vulnerability indexes, Water (Switzerland), vol. 13, no. 1, 2021, doi: 10.3390/w13010098 DOI: https://doi.org/10.3390/w13010098
[19] M. Przybya-Kasperek and K. F. Marfo, A multi-layer perceptron neural network for varied conditional attributes in tabular dispersed data, 2024, Public Library of Science (PLoS). doi: 10.1371/journal.pone.0311041 DOI: https://doi.org/10.1371/journal.pone.0311041
[20] Q. Zheng et al., Deep Tabular Data Modeling with Dual-Route Structure-Adaptive Graph Networks, IEEE Trans. Knowl. Data Eng., vol. 35, no. 9, pp. 97159727, 2023, doi: 10.1109/TKDE.2023.3249186 DOI: https://doi.org/10.1109/TKDE.2023.3249186
[21] N. Shazeer, GLU Variants Improve Transformer, no. 4, pp. 15, 2020, [Online]. Available: http://arxiv.org/abs/2002.05202
[22] Y. Sun, S. Brockhauser, and P. Hegedus, Comparing End-to-End Machine Learning Methods for Spectra Classification, Appl. Sci., vol. 11, no. 23, 2021, doi: 10.3390/app112311520 DOI: https://doi.org/10.3390/app112311520
[23] J. Si, W. Y. Cheng, M. Cooper, and R. G. Krishnan, InterpreTabNet: Distilling Predictive Signals from Tabular Data by Salient Feature Interpretation, in Proceedings of Machine Learning Research, S. R., K. Z., H. K., W. A., O. N., S. J., and B. F., Eds., University of Toronto, Canada: ML Research Press, 2024, pp. 4535345405.
[24] Q. W. Khan et al., Predictive Modeling of Water Table Depth, Drilling Duration, and Soil Layer Classification Using Adaptive Ensemble Learning for Cost-Effective Percussion Water Borehole Drilling, IEEE Access, vol. 11, pp. 7670376721, 2023, doi: 10.1109/ACCESS.2023.3298225 DOI: https://doi.org/10.1109/ACCESS.2023.3298225
[25] C. Yu et al., Gradient Boosting Decision Tree with LSTM for Investment Prediction, Proc. - 2025 5th Asia-Pacific Conf. Commun. Technol. Comput. Sci. ACCTCS 2025, pp. 5762, 2025, doi: 10.1109/ACCTCS66275.2025.00017 DOI: https://doi.org/10.1109/ACCTCS66275.2025.00017
[26] J. Sherozbek, J. Park, M. S. Akhtar, and O.-B. Yang, Transformers-Based Encoder Model for Forecasting Hourly Power Output of Transparent Photovoltaic Module Systems, Energies, vol. 16, no. 3, 2023, doi: 10.3390/en16031353 DOI: https://doi.org/10.3390/en16031353
[27] T. Ebbs-Picken, D. A. Romero, C. M. Da Silva, and C. H. Amon, Deep encoderdecoder hierarchical convolutional neural networks for conjugate heat transfer surrogate modeling, Appl. Energy, vol. 372, 2024, doi: 10.1016/j.apenergy.2024.123723 DOI: https://doi.org/10.1016/j.apenergy.2024.123723
[28] N. Koenen and M. N. Wright, Interpreting Deep Neural Networks with the Package innsight, J. Stat. Softw., vol. 111, no. 8, 2024, doi: 10.18637/jss.v111.i08 DOI: https://doi.org/10.18637/jss.v111.i08
[29] A. A. Novita, R. Ngindana, E. Putra, and D. Virgiyansha, Development and challenges in the implementation of sustainable development goals (SDGs) in Indonesia: A systematic literature review, J. Inov. Ilmu Sos. dan Polit., vol. 5, no. 2, pp. 189196, 2023, doi: 10.33474/jisop.v5i2.21192 DOI: https://doi.org/10.33474/jisop.v5i2.21192
[30] M. N. P. Ramadhan, The Role of SDG 4: Quality Education on the Internationalization, Hasanuddin J. Strateg. Int. Stud., vol. 1394, no. 2, pp. 3951, 2023, doi: 10.20956/hjsis.v1i2.27442 DOI: https://doi.org/10.20956/hjsis.v1i2.27442
[31] M. C. Mayer, M. B. Zafar, L. Franceschi, and H. Rangwala, Hands-on Tutorial: explanations in AI: Methods, Stakeholders and Pitfalls, in Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Amazon Web Services, Seattle, WA, United States: Association for Computing Machinery, 2023, pp. 57835785. doi: 10.1145/3580305.3599181 DOI: https://doi.org/10.1145/3580305.3599181
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Harfely Leipary, Suryasatria Trihandaru, Hanna Arini Parhusip
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
How to Cite
Share
Plum Analytics