Optimizing 1D Clustering Through Customization is a blog post by Ido Hirsh and AppsFlyer Engineering. In order to obtain the optimal solution for our case, we customize a one-dimensional clustering algorithm to suit our needs. We discuss the challenges we face while customizing off-the-shelf algorithms.
Table of contents
Optimizing 1D Clustering Through CustomizationProblem contextClassic clustering adaptationPutting business logic into the modelCentroid Calculation’s Effects on ClusteringExperimentLessons learned18 Impressions