STA 555 REVISION 1 AND 2

In directed or supervised Data Mining, the data set is divided into testing and validation set

• true

• false

Answer explanation

the correct anwer is : in directed or supervised data mining the data set is divided into TRAINING and VALIDATION SET

Dataset that has outliers can cause variances in results and can impact our analysis outcomes

• true

• false

Predictive rules or estimated equations are established based on the training data while the best model is determined based on the validated data

• TRUE

• FALSE

Higher dimensionality of the data would always decrease the misclassification rate

• TRUE

• FALSE

Answer explanation

Higher dimensionality of the data decrese accurancy

When a node has only one class, its score is 0 when applying the Gini splitting algorithm

• false

• true

Answer explanation

the score is 1 not 0

choices of splitting algorithm are always depending on the type of target variable

• true

• false

best model needs to have smallest error and highest ROC index

• true

• false

___________ is a data mining technique that does not require the missing values to be imputed in order to build predictive model

Variable selection is the process of selecting a subset of the available inputs that are useful for predicting the target variable

• True

• False

If the imbalance sample issue is not solved the estimated model will likely predict the rare event better than the non rare event

• True

• False

Answer explanation

Rare tu macam minority dan non rare tu macam majority jadi obviously non rare event better that rare event