Clickstream Cluster Visualization Using UMAP and DBSCAN
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing patron study and survey methods for analyzing human interactions are limited in scope, costly, and rely on known hypotheses, failing to capture unknown behaviors effectively.
Innovation Solution
An apparatus and method utilizing unsupervised machine learning algorithms, including UMAP, DBSCAN, and KNN, to analyze patron clickstream data and graphically illustrate clusters without relying on known questions or hypotheses, using non-linear dimensionality reduction and clustering algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional survey methods are used to analyze patron interactions, then the analysis is based on known hypotheses and categories, but the scope is limited, cost is high, and results are biased due to participant selection
Solution Approach 1:
The system performs unsupervised learning automatically without human intervention in defining categories or hypotheses. The algorithm independently processes raw clickstream data, automatically identifying patterns and segments, thereby eliminating survey design biases and expanding analytical scope to include unknown behaviors
Solution Approach 2:
The patent replaces the mechanical survey process (designing questions, recruiting participants, collecting responses) with an automated computational system that processes digital clickstream data. This substitution eliminates human bias in question formulation and participant selection while expanding the scope to all patrons who leave digital traces
2Adaptability or versatility
If unsupervised machine learning algorithms are applied to high-dimensional clickstream data, then the analysis scope is expanded and cost is reduced, but the complexity of data processing increases
Solution Approach 1:
The patent segments the complex high-dimensional clickstream data into manageable components by applying dimensionality reduction techniques. The system divides the processing into distinct stages: data collection, dimensionality reduction, clustering, and visualization, making the overall complex task tractable through systematic breakdown
Solution Approach 2:
The patent transforms high-dimensional clickstream data into lower-dimensional representations using techniques like t-SNE or PCA. This dimensionality reduction converts complex multi-dimensional patron behavior data into 2D or 3D visualizations that can be easily interpreted while preserving essential patterns and relationships
3Loss of information
If raw clickstream data is processed without dimensionality reduction, then the analysis maintains full detail, but the computational cost and processing time increase significantly
Solution Approach 1:
The patent applies dimensionality reduction that retains sufficient information for meaningful analysis while discarding redundant dimensions. The technique preserves the most informative aspects of patron behavior patterns while reducing computational burden, achieving a balance between maintaining data fidelity and improving processing efficiency
Data Source
AI summary
Groups of patrons may be discovered by measuring website and mobile site patron clickstream data in a mathematical and unsupervised way over a predetermined time and by graphically clustering the patron clickstream data using non-linear dimensionality reduction in the form of a Uniform Manifold Approximation and Projection algorithm (UMAP). The data from the UMAP may then be feed into a Density Based Spatial Clustering of Applications with Noise algorithm (DBSCAN) in order to identify a center of each cluster. Next, using the data from the UMAP and the center of each cluster from the DBSCAN, a K-Nearest Neighbor algorithm (KNN) may be applied to identify data points closest to the center of each cluster and to shade each of the data points to graphically identify each cluster of the plurality of clusters. Next, illustrate a graph on the display representative of the data points shaded following application of the KNN.