Latent student clustering using a hierarchical block clustering method

Abstract

The present disclosure relates to a system for verifying homogeneity in clusters and comprises a processor and a non-transitory computer readable memory storing instructions that are executable by the processor. The system creates at least one adjacency matrix representing a relationship between rows and columns of a table. The system applies an algorithm to the table to identify a first set of clusters. For each identified cluster, the system compares a computed dispersion value to a predetermined threshold. The cluster is associated as being homogeneous if the threshold is not met and not homogeneous if the threshold is met. The system reapplies the algorithm to the set of non-homogenous clusters within the data set and repeats the thresholding processes until each cluster resulting from the data set is homogeneous. Upon reaching the desired homogeneity for each cluster, the system recombines the clusters sequentially while keeping the each identified cluster together.