25th EANN 2024, 27 - 30 June 2024, Corfu, Greece

Machine Learning Classification of Water Conductivity raw values of "Faneromeni" Reservoir in Crete

Lazaros Iliadis, Nichat Kiourt, Christos Akratos, Antonios Papaleonidas

Abstract:

  Water Conductivity is a measure of dissolved salts’ concentration in the water. It depends on the concentration of ions and on water temperature. Assessment of the conductivity values is a priority, as the higher its values the more dangerous it becomes for humans. Therefore, the first purpose of this research is to evaluate and classify the water conductivity levels in the "Faneromeni" reservoir located in Crete. This was achieved by developing powerful Machine Learning models. The raw data from the survey area comprised of simple crisp conductivity measurements, which were successfully converted to labels by developing Quartiles, during preprocessing. This was followed by the development of Machine Learning models, which successfully yield four labels 'Low', 'Medium', 'High' and 'Extreme' assigned to each record. Subsequently, a comparative discussion was performed (for the first time in the literature) between the obtained outcome with the relative one from our previous research effort, concerning the reservoir of the 'Bramianon' area of Crete.  

*** Title, author list and abstract as seen in the Camera-Ready version of the paper that was provided to Conference Committee. Small changes that may have occurred during processing by Springer may not appear in this window.