Implementation:DistrictDataLabs Yellowbrick WordCorrelationPlot

Knowledge Sources	DistrictDataLabs_Yellowbrick
Domains	NLP, Visualization
Last Updated	2026-02-08 05:00 GMT

Overview

Concrete tool for visualizing word co-occurrence correlation as a heatmap across a text corpus, provided by the Yellowbrick text module.

Description

The WordCorrelationPlot computes pairwise Pearson correlation coefficients between specified words based on their co-occurrence patterns in documents. It renders a heatmap with color-coded correlation values and optional colorbar. The underlying computation uses scikit-learn's CountVectorizer to build term-document matrices.

Usage

Import this visualizer when analyzing relationships between specific words in a text corpus. It is useful for understanding which terms tend to appear together in documents.

Code Reference

Source Location

• Repository: DistrictDataLabs_Yellowbrick
• File: yellowbrick/text/correlation.py
• Lines: 1-349

Signature

class WordCorrelationPlot(TextVisualizer):
    def __init__(
        self,
        words,
        ignore_case=False,
        ax=None,
        cmap="RdYlBu",
        colorbar=True,
        fontsize=None,
        **kwargs,
    ):
        """Visualizes word correlation as a heatmap."""

def word_correlation(
    words, corpus, ignore_case=True, ax=None, cmap="RdYlBu",
    show=True, colorbar=True, fontsize=None, **kwargs,
):
    """Quick method for one-off word correlation visualization."""

Import

from yellowbrick.text import WordCorrelationPlot
from yellowbrick.text.correlation import word_correlation

I/O Contract

Inputs

Name	Type	Required	Description
words	list of str	Yes	Words to compute correlations for
X	list of str	Yes	Corpus of documents (fit)
ignore_case	bool	No	Case-insensitive matching (default: False)
cmap	str	No	Colormap for heatmap (default: "RdYlBu")

Outputs

Name	Type	Description
ax	matplotlib.Axes	Axes with correlation heatmap

Usage Examples

from yellowbrick.text import WordCorrelationPlot
from yellowbrick.datasets import load_hobbies

corpus = load_hobbies()

words = ["game", "sport", "music", "movie"]
viz = WordCorrelationPlot(words)
viz.fit(corpus.data)
viz.show()

Related Pages

• Environment:DistrictDataLabs_Yellowbrick_Python_Scikit_Learn_Environment