Exam DP-100 topic 7 question 9 discussion

You must prioritize the columns of data for predicting the outcome(The mutual information score measures the contribution of a variable towards reducing uncertainty about the value of another variable: namely, the label). You must use non-parametric statistics to measure relationships. i think asnwer is Mutual information

Fisher test is better, it is similar to Chi squared but fisher is more exact, choose Fisher score

I think the answer is Fisher Score Fisher Score: Label can be text or numeric but features must be numeric. Mutual Information: Labels and features can be text or numeric. Use this method for computing feature importance for two categorical columns. Chi Squared: Labels and features can be text or numeric. Use this method for computing feature importance for two categorical columns.

Since both MedianValue and Averagenumber of house is numerical variable, we should use Fisher Price

The answer should be Fisher Score since the label and the features are numeric values. Chi Squared and Mutual information for categorical values.

By the definition of Mutual Information, a low value should mean that one feature does not give me information about the other and by the definition of Chi Square, a low value of Chi Square means that the two features must be independent. Hence i guess Chi square is not used and mutual information is used

The answer is counts. Chi square and mutual information applies to categorical data. Since there is only 1 feature we looking at meaning there's only one dataset it cannot be Fisher. https://docs.microsoft.com/en-us/azure/machine-learning/studio-module-reference/filter-based-feature-selection