Exam DP-100 topic 3 question 30 discussion

I think the answer should be C. Chi-squared as both label and features are categorical.

C is the answer. Your choice of a filter selection method depends in part on what sort of input data you have. The requirement for all Pearson Correlation, Spearman Correlation and Fisher Score methods is "features must be numeric". But for Chi Squared, the requirement is "features can be text or numeric" so you can use this method for computing feature importance for categorical columns. See the table at this link: https://docs.microsoft.com/en-us/azure/machine-learning/studio-module-reference/filter-based-feature-selection

I think the correct is C too

The best statistical method to use for filter-based feature selection in this multi-class classification scenario with categorical features is Chi-squared. The chi-squared test measures dependence between categorical variables. It will identify categorical features that have a statistically significant correlation with the label column.

Its fun because in this answer explanation is stated: "Pearson's correlation coefficient is the test statistics that measures the statistical relationship, or association, between two __continuous variables__." so the answer can't be Pearson ... !!!

C - ChatGPT

Chi-sqaure works for categorical data the rest don't so it should be the answer

When features as well as label are categorical values we can use chi-squared method

C is correct

C is the answer.

Those "correct answers" were not made by data scientists for sure.

Chi-squared. It's categorial.

The answer should be C. Chi-Squared Feature -> Numeric | Text Label -> Numeric | Text Pearson | Kendall | Spearman | Fisher Score Feature -> Numeric Label -> Numeric | Text https://docs.microsoft.com/en-us/azure/machine-learning/studio-module-reference/filter-based-feature-selection

Agree with C

agree: Chi-squared

the answer is C

Answer is definitely C: Chi Squared is used in Categorical features and . Pearson is used for continuous not categorical