We have a UI for visualizing pump and weather data as well as some basic statistical processing of these datasets. This allowed us to do some exploratory date analysis to manually look for patterns in the data and overall get a grip of how to proceed with more advanced/sophisticated ML methods.
From the work mentioned above, it is clear that there are a lot of variables to consider with perhaps complex relationships when trying to explain the observed numbers in the graphs. The ituition is that using ML methods gives us a computationally feasable way to discover patterns in the data and to perform prediction.
User stories
In the UI, would like to manually type in the values of some parameters (like e.g. the precipitation level for the past 3 days) and predict datapoints for the specified timeperiod. -> can compare with actual values to see effect of variables
Considerations
How could the model capture the following scenario: During the december it snowed a lot, this precipitation is not immidiatly felt in the pump stations, but only after the temprature has risen above som level for an extended period of time (melting snow)
We have a UI for visualizing pump and weather data as well as some basic statistical processing of these datasets. This allowed us to do some exploratory date analysis to manually look for patterns in the data and overall get a grip of how to proceed with more advanced/sophisticated ML methods.
From the work mentioned above, it is clear that there are a lot of variables to consider with perhaps complex relationships when trying to explain the observed numbers in the graphs. The ituition is that using ML methods gives us a computationally feasable way to discover patterns in the data and to perform prediction.
User stories
Considerations