Exam Certified Machine Learning Professional topic 1 question 13 discussion

Categorical Features ->Summary Statistics: Mode, Number of unique values, Number of missing values

C is the correct answer

Among the options, the most appropriate simple statistic to monitor for categorical feature drift is mode. The others aren't right due to the below reasons: Number of unique values - while useful for understanding the diversity of categories, it does not directory indicate drive unless the number of categories changes significantly. Percentage of missing values - while important for data quality, it does not directly indicate drift in the distribution of categorical values.

would go with C