Implementation of human–computer interaction (HCI) in flash flood prediction system using machine learning algorithm
Keywords:
Human computer interaction, Machine learning, Flash flood prediction, Information systems, Usability testingAbstract
Flash floods are a hydrometeorological disaster that significantly impact human safety and economic losses. An accurate and user-friendly prediction system is essential for disaster mitigation. This study aims to develop a flash flood prediction system that integrates machine learning algorithms with Human–Computer Interaction (HCI) principles to improve prediction accuracy and user experience. Hydrological and meteorological data including rainfall, air humidity, water level, and flow velocity were used as input variables for Random Forest and Support Vector Machine machine learning models. Performance testing was conducted by comparing accuracy, precision, recall, and F1-score. Furthermore, HCI evaluation was conducted through usability testing using parameters such as efficiency, effectiveness, and user satisfaction. The results showed that the integration of machine learning was able to provide high-accuracy flash flood predictions, while the application of HCI principles to the mobile and web interfaces improved ease of navigation, information clarity, and system responsiveness. Users reported a significant increase in perceived usability and comfort of the system based on the System Usability Scale (SUS) results, with an average score in the “Good” category. Thus, this research contributes to the development of a disaster prediction system that is not only technically reliable but also easy to operate by the public and stakeholders in disaster mitigation decision-making.
References
1. Al-Maliki, A. A., & Rahman, M. M. (2021). Flood hazard prediction using machine learning and hydrological variables: A comparative study. Natural Hazards Review, 22(3), 04021011. https://doi.org/10.1061/(ASCE)NH.1527-6996.0000473
2. Aryal, K. R., & Thapa, S. (2020). Time-series rainfall forecasting using LSTM networks in monsoon-dominated regions. Journal of Water and Climate Change, 11(4), 1560–1572. https://doi.org/10.2166/wcc.2020.092
3. BMKG. (2023). Data Curah Hujan Harian Nasional Tahun 2010–2023. Badan Meteorologi, Klimatologi, dan Geofisika. https://www.bmkg.go.id
4. BNPB. (2023). Data Kejadian Bencana Banjir dan Banjir Bandang di Indonesia Periode 2000–2023. Badan Nasional Penanggulangan Bencana. https://www.bnpb.go.id
5. Bui, D. T., & Hoang, N. D. (2020). Flood susceptibility prediction using Random Forest and GIS-based hydrological features. Arabian Journal of Geosciences, 13(21), 1120. https://doi.org/10.1007/s12517-020-06009-3
6. Chen, T., & Koo, B. (2021). Enhancing rainfall–runoff forecasting using LSTM for hydrological time-series data. Water Supply, 21(8), 4253–4265. https://doi.org/10.2166/ws.2021.147
7. Chen, Y., & Wu, H. (2021). Long Short-Term Memory Networks for Rainfall and Flood Forecasting in Hydrometeorological Applications. Journal of Hydrology and Environmental Systems, 48(2), 115–129. https://doi.org/10.1016/j.hydes.2021.115129
8. Das, S., & Mohanty, S. (2022). Human-centered flood warning interface design to reduce cognitive load during emergencies. International Journal of Human–Computer Studies, 162, 102800. https://doi.org/10.1016/j.ijhcs.2022.102800
9. Dewi, R., & Hasanah, N. (2021). Usability evaluation of disaster alert mobile applications using USE Questionnaire. Indonesian Journal of Electrical Engineering and Computer Science, 24(1), 411–418. https://doi.org/10.11591/ijeecs.v24.i1.pp411-418
10. Ghosh, S., & Samanta, S. (2020). Flood forecasting using machine learning models: Performance comparison and feature importance. Modeling Earth Systems and Environment, 6(3), 1579–1591. https://doi.org/10.1007/s40808-020-00786-1
11. Hidayat, M., & Kurniawan, A. (2022). Mobile early warning systems for hydrometeorological hazards: A usability-focused evaluation. Journal of Theoretical and Applied Information Technology, 100(2), 350–361.
12. Islam, M. A., & Pradhan, B. (2021). Machine learning-based flood risk mapping using hydrometeorological and topographic datasets. Hydrological Sciences Journal, 66(12), 1680–1694. https://doi.org/10.1080/02626667.2021.1957919
13. Kang, J. S., Park, S. H., & Min, K. H. (2021). Challenges of Meteorological Big Data for Flood Early Warning Systems: Spatial and Temporal Data Limitations. International Journal of Disaster Risk Science, 12(4), 589–603. https://doi.org/10.1007/s13753-021-00367-4
14. Koushik, S. M., & Verma, A. (2023). Real-time LSTM modeling for rainfall prediction in tropical regions. Environmental Challenges, 10, 100658. https://doi.org/10.1016/j.envc.2023.100658
15. Lee, M. J., Yoon, S. J., & Cho, H. R. (2022). Improving Disaster Warning Interfaces through Color-Coded Risk Visualization and Cognitive Load Reduction Techniques. International Journal of Human–Computer Studies, 165, 102848. https://doi.org/10.1016/j.ijhcs.2022.102848
16. Liu, H., & Zhang, Q. (2021). User-centered visualization for flood risk communication using color-coded alerts. Journal of Flood Risk Management, 14(3), e12721. https://doi.org/10.1111/jfr3.12721
17. Mulyani, N., & Setiawan, A. (2022). Integration of IoT rainfall sensors and machine learning for localized flood detection. International Journal of Intelligent Engineering and Systems, 15(6), 450–460. https://doi.org/10.22266/ijies2022.1231.39
18. Nielsen, J. (2010). Usability Engineering. Morgan Kaufmann.
19. Norman, D. A. (2013). The Design of Everyday Things: Revised and Expanded Edition. Basic Books.
20. Rahman, A. F., Nugroho, Y. A., & Dewi, R. P. (2022). Usability Assessment of Mobile-Based Disaster Early Warning System Using the USE Questionnaire. Journal of Information Technology and Human Interaction, 7(1), 45–58. https://doi.org/10.30595/jithu.v7i1.2022
21. Rahman, M. F., & Alam, M. (2022). Performance evaluation of deep learning models for hydrological forecasting in complex terrains. Water Supply, 22(3), 3440–3454. https://doi.org/10.2166/ws.2022.051
22. Santos, L. R., & Silva, M. M. (2023). A cognitive load–aware interface for emergency disaster warnings. Computers in Human Behavior Reports, 9, 100230. https://doi.org/10.1016/j.chbr.2022.100230
23. Sari, M. P., & Pratama, W. D. (2021). The Impact of Cognitive Load-Oriented Interface Design on User Understanding in Flood-Prone Areas. Indonesian Journal of Human–Computer Interaction, 3(2), 78–90. https://doi.org/10.24123/ijhci.v3i2.4587
24. Yunanto, R., & Putra, H. (2021). Evaluation of mobile disaster mitigation systems using HCI principles. Journal of Information and Communication Technology, 20(4), 489–507. https://doi.org/10.32890/jict2021.20.4.2
25. Zhang, L., Chen, Z., & Liu, Y. (2020). Application of Deep Learning Models for Meteorological Time-Series Prediction: A Case Study Using LSTM Networks. Atmospheric Research, 234, 104743. https://doi.org/10.1016/j.atmosres.2020.104743
Downloads
Published
Conference Proceedings Volume
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.