MLS-C01 Question 112
Single answerYou are a data scientist developing an ML model to predict customer churn using Amazon SageMaker. You have trained your model and now want to visualize the distribution of predicted probabilities for churn (ranging from 0 to 1) and identify any anomalies. Which type of graph should you use to achieve this?
- A
Scatter plot
- B
Histogram
- C
Box plot
- D
Time series plot
Show answer and explanation
Correct answer: B
Explanation
To analyze the distribution of predicted probabilities for churn (a single variable), a histogram is the most appropriate visualization technique. It allows you to observe how the probabilities are spread across their range and identify any anomalies, such as unusual peaks or gaps. Other plot types, like scatter plots and box plots, are less suitable for this purpose, and time series plots are irrelevant as the data is not time-dependent.
- A. Incorrect.
Scatter plots are used to visualize relationships between two variables, but they are not ideal for visualizing the distribution of a single variable.
- B. Correct.
Histograms are specifically designed to visualize the distribution of a single variable by grouping values into bins. This makes it the ideal choice for analyzing predicted probabilities.
- C. Incorrect.
Box plots are useful for summarizing the distribution of a variable, including its median and quartiles, but they are less effective than histograms for identifying patterns or anomalies in the distribution of a continuous variable like predicted probabilities.
- D. Incorrect.
Time series plots are used to analyze data points over time, but this scenario does not involve time-dependent data.