Classification

The ratio of correctly predicted positive observations is known as:

• Precision

• Accuracy

• Bias

• Recall

What is the purpose of a confusion matrix?

• Helps to select the best model

• Calculate precision and accuracy

• Table that describes the performance of a classification model

• To generate a classification report

Which elements will you see in a classification report?

• mean squared error

• recall

• R2 value

• accuracy

Which models use weak learners?

• Random Forests

• Decision Tree

• XGBoost

• Support Vector Machines

Which functions might we use for preprocessing categorical data?

• get_dummies()

• to_string()

• StandardScaler()

• LabelEncoder()

What can you learn from the decision tree visualization?

• How many nodes and branches the model has.

• How accurate the model is.

• Whether the model might be overfitting.

• How deep the tree goes.

How is recall defined?

• The total percent of accurate predictions

• The proportion of predicted positives that are accurate

• The proportion of actual positives that are accurate

• The proportion of true positives to true negatives

Which is undersampling?

• Taking fewer of the majority class

• Taking more of the minority class

• Adding data points averaged from both classes

• Pruning a long branch of a decision tree

How do you generally interpret a PR curve?

• The model with the highest data point is the best

• The model with the smoothest curve is the best

• The model with the least variation in the curve is the best

• The model with the most area under the curve is the best

When a search engine returns 30 pages, only 20 of which are relevant, while failing to return 40 additional relevant pages, it's precision and recall are

• its precision is 20/40, while its recall is 20/30

• its precision is 20/30, while its recall is 40/30

• its precision is 20/30, while its recall is 20/60