The computationally expensive fitting procedure would be applied to every event, instead of the few percent of the events that are of interest for the analysis. An unsupervised ML algorithm able to separate the data without a priori knowledge of the different types of events increases the efficiency of the analysis tremendously, and allows the downstream analysis to concentrate on the fitting efforts only on events of interest. In addition, the clustering allows for more exploration of the data, potentially enabling new discovery of unexpected reaction types.