Fig. 8: Statistical analysis of the prediction goodness for the first six rearrangement events from the test set.

The y-axis represents the prediction goodness, \({{{\mathcal{C}}}}_{{\tau }_{c}}(n)\), for each given event n. The lines joining the means of all the boxplots showcase the average prediction goodness of each neural network with respect to the sequence of the events. The box represents the spread of data between quartiles 1 and 3, while the whiskers extend from the box to the minimum and maximum values of the data.