k-Box Plot
This component generates two powerful box plot visualizations for a selected numerical column: a Standard Box Plot and a Modified Box Plot. In addition to the plots, the output port provides a detailed summary table containing the five-number summary along with the Lower bound, Upper bound, Total count, Outlier count, and Outlier percentage.
The Standard Box Plot displays the full data range, with whiskers extending to the absolute minimum and maximum values. The Modified Box Plot explicitly identifies outliers based on the Interquartile Range (IQR).
The primary feature of this component is its flexibility. Unlike other available nodes, it allows users to configure the k multiplier to define the outlier detection threshold (default is 1.5). This introduces a novel capability to the KNIME community, enabling a more nuanced analysis tailored to specific datasets or domain requirements.
Configuration Options:
k multiplier: Set a custom value for the IQR rule to adjust outlier sensitivity.
Column selection: Choose the single numerical column for analysis.
Deactivate Plots for Faster Running: An optional setting to disable plot generation. This significantly improves performance, especially during iterative executions inside loops, when only the summary output table is required.
Version 3.0
Created by: Carlos Enrique Diaz, MBM, B.Eng.
Email: carlos.diaz@usask.ca