Univariate Feature Selection

This notebook is an example of using univariate feature selection to improve classification accuracy on a noisy dataset.

In this example, some noisy (non informative) features are added to the iris dataset. Support vector machine (SVM) is used to classify the dataset both before and after applying univariate feature selection. For each feature, we plot the p-values for the univariate feature selection and the corresponding weights of SVMs. With this, we will compare model accuracy and examine the impact of univariate feature selection on model weights.

# Authors: The scikit-learn developers
# SPDX-License-Identifier: BSD-3-Clause

Generate sample data

import numpy as np

from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split

# The iris dataset
X, y = load_iris(return_X_y=True)

# Some noisy data not correlated
E = np.random.RandomState(42).uniform(0, 0.1, size=(X.shape[0], 20))

# Add the noisy data to the informative features
X = np.hstack((X, E))

# Split dataset to select feature and evaluate the classifier
X_train, X_test, y_train, y_test = train_test_split(X, y, stratify=y, random_state=0)

Univariate feature selection

Univariate feature selection with F-test for feature scoring. We use the default selection function to select the four most significant features.

from sklearn.feature_selection import SelectKBest, f_classif

selector = SelectKBest(f_classif, k=4)
selector.fit(X_train, y_train)
scores = -np.log10(selector.pvalues_)
scores /= scores.max()

import matplotlib.pyplot as plt

X_indices = np.arange(X.shape[-1])
plt.figure(1)
plt.clf()
plt.bar(X_indices - 0.05, scores, width=0.2)
plt.title("Feature univariate score")
plt.xlabel("Feature number")
plt.ylabel(r"Univariate score ($-Log(p_{value})$)")
plt.show()