An initiative of HKV line in water and Rijkswaterstaat
The use of machine learning in the water sector has increased significantly in recent years and offers opportunities to solve important questions in the water sector. At the same time, many people question the reliability of the outcomes How should you trust a model that is not transparent in how it makes calculations? The DigiShape seedmoney 2023 project “Explainable AI” aims to change this.
A project on salt intrusion being conducted at the Department of Public Works discovered a number of ways to make machine learning prediction more insightful. Zo is het belangrijk om niet alleen de uitkomsten, maar ook de manier van rekenen te visualiseren. For example, it is important to visualize not only the outcomes, but also the method of computation. We use Explainable AI techniques so you can see which key factors are most important. You can also, for example, change one variable at a time and see how it affects the outcomes. In this way, you discover what influence a key factor has on the outcomes of the model. By displaying this in graphs, you get more insight into what is happening in the black box. In this seedmoney project, further methods of providing insight are being explored. Based on these techniques, a tool was developed that allows users to quickly and interactively go through the results of a model to see which key factors are most important. For this, you then do not need any knowledge of AI yourself.
More information in the interview with Paula Lambregts and Thomas Stolp from HKV.
Report and code
- These are in development and will be posted on this page as soon as they are ready.
