Answer-first summary for fast verification
Answer: Chi-squared
The question involves filter-based feature selection for categorical features that are highly correlated with a categorical label column in a multi-class classification scenario. The community discussion overwhelmingly supports Chi-squared (C) as the correct answer, with 100% of answers selecting C and numerous comments providing detailed reasoning. Key points from the discussion include: Chi-squared is specifically designed for categorical-categorical relationships, while Pearson, Kendall, and Spearman correlations require numeric features. Microsoft's official documentation for filter-based feature selection in Azure ML confirms that Chi-squared supports both text and numeric features and is appropriate for computing feature importance between two categorical columns. The other options (A, B, D) are unsuitable because they are correlation methods designed for continuous/numeric variables, not categorical data.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
You are performing filter-based feature selection on a dataset to build a multi-class classifier using Azure Machine Learning Studio. The dataset contains categorical features that are highly correlated with the label column.
You need to select the appropriate feature scoring statistical method to identify the key predictors.
Which method should you use?
A
Kendall correlation
B
Spearman correlation
C
Chi-squared
D
Pearson correlation
No comments yet.