AIP-210 試験問題を無料オンラインアクセス

Your dependent variable data is a proportion. The observed range of your data is 0.01 to 0.99. The instrument used to generate the dependent variable data is known to generate low quality data for values close to 0 and close to 1. A colleague suggests performing a logit-transformation on the data prior to performing a linear regression. Which of the following is a concern with this approach?
Definition of logit-transformation
If p is the proportion: logit(p)=log(p/(l-p))

• A.Noisy data could become more influential in your model.
• B.After logit-transformation, the data may violate the assumption of independence.
• C.The model will be more likely to violate the assumption of normality.
• D.Values near 0.5 before logit-transformation will be near 0 after.

You have a dataset with many features that you are using to classify a dependent variable. Because the sample size is small, you are worried about overfitting. Which algorithm is ideal to prevent overfitting?

• A.Random forest
• B.Decision tree
• C.XGBoost
• D.Logistic regression

Which of the following metrics is being captured when performing principal component analysis?

• A.Skewness
• B.Kurtosis
• C.Missingness
• D.Variance

The graph is an elbow plot showing the inertia or within-cluster sum of squares on the y-axis and number of clusters (also called K) on the x-axis, denoting the change in inertia as the clusters change using k-means algorithm.
What would be an optimal value of K to ensure a good number of clusters?

• A.5
• B.3
• C.2
• D.9

Which of the following statements are true regarding highly interpretable models? (Select two.)

• A.They usually compromise on model accuracy for the sake of interpretability.
• B.They are usually referred to as "black box" models.
• C.They are usually very good at solving non-linear problems.
• D.They are usually easier to explain to business stakeholders.
• E.They are usually binary classifiers.