Unsupervised apparatus and method for graphically clustering high dimensional patron clickstream data

A method for graphically clustering high dimensional patron clickstream data comprising:

• receiving and storing, by a processor, patron clickstream data over a predetermined time period;
• determining, by the processor, the predetermined time period elapsed;
• transforming, by the processor, the patron clickstream data into a probability matrix using a Uniform Manifold Approximation and Projection (UMAP) algorithm, where the probability matrix comprises a probability that a patron would proceed from a first page to a second page;
• transforming, by the processor, the probability matrix by reducing dimensionality into two dimensional data using the UMAP algorithm and generating, by the processor, a cluster graph visualizing a plurality of clusters;
• identifying, by the processor using a Density Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm applied to the two dimensional data, a plurality of clusters and the center of each cluster of the plurality of clusters;
• applying, by the processor, a K-Nearest Neighbor algorithm (KNN) algorithm to identify data points closest to the center of each cluster;
• and
• shading, by the processor, each of the identified data points to graphically identify each cluster of the plurality of clusters;
• and
• outputting, by the processor on a display, a graph representative of the data points shaded following application of the KNN algorithm.