Prediction of a Dam’s Hazard Level

A Case Study from South Africa
  • Urna Kundu
  • Srabanti Ghosh
  • Satyakama PaulEmail author
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 906)


South Africa has a vast infrastructure of dams. Since the country receives very little rainfall, these dams assume prime importance in storing water and sustaining agriculture, industry, household, etc. Thus prediction of their multiple hazard levels (in this case, three) is of prime importance. In addition, South Africa lacks skilled personnels to classify these dam’s hazards. Under such a framework, this work is an application of single and ensemble decision trees in a multi-class supervised learning framework to predict the hazard level of a dam. The result obtained is highly promising and at is above 94\(\%\). With the implementation of the algorithm, we expect to address the problem of paucity of skilled personnels.


Dam hazard-level prediction Multiclass classification Imbalanced classes Decision trees C5.0 Tree bagging Random forest t-SNE South Africa 


  1. 1.
    National Water Act, Act No 36 of 1998,
  2. 2.
    Dam Sefety Office, Dept. of Water and Sanitation - Republic of South Africa,
  3. 3.
    The World Bank, Average precipitation in depth (mm per year), World Bank,
  4. 4.
    Dam safty - ensuring the interity of SA’s 5000+ registered dams, South African Water Research Commission.
  5. 5.
  6. 6.
  7. 7.
    Hartford, D. N. D., & Baecher, G. B. (2004). Risk and Uncertainty in Dam Safety. Inst of Civil Engineers Pub. ISBN-13:978-0727736390Google Scholar
  8. 8.
    Danso-Amoako, E., Scholz, M., Kalimeris, N., Yang, Q., & Shao, J. (2012). Predicting dam failure risk for sustainable flood retention basins: A generic case study for the wider Greater Manchester area. In Computers, Environment and Urban Systems (Vol. 36, pp. 423–433).Google Scholar
  9. 9.
    Maaten, L., & Hilton, G. (2008). Visualizing Data using t-SNE. Journal of Machine Learning Research, 9, 2579–2605.zbMATHGoogle Scholar
  10. 10.
    Kohavi, R., & John, G. H. (1997). Wrappers for feature subset selection . In Applied Intelligence (pp. 273–324).Google Scholar
  11. 11.
    Kursa, M. B., & Rudnicki, W. R. (2010). Feature selection with the boruta package. Journal of Statistical Software, 36(11), 1–13.CrossRefGoogle Scholar
  12. 12.
    Guyon, I., Weston, J., & Barnhill, S. (2002). Gene selection for cancer classification using support vector machines. In Machine Learning (Vol. 46, pp. 389–422).Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  1. 1.WNSBengaluruIndia
  2. 2.Oracle, Prestige Tech ParkBengaluruIndia

Personalised recommendations