MODULE 2 S2.3.1

Question 1

T/F In Naïve Bayes, training and testing are fast and easy to understand and process.

Answer 1

TRUE

Question 2

3 Kinds of Naïve Bayes Classifier in Scikit-learn

Answer 2

GaussianNB - continuous data
BernoulliNB - binary data, text data
MultinomialNB - integer count data, text data

Question 3

Classification algorithms in linear models

Answer 3

• How well weights and intercepts fir the training data
• Cost function or loss function
• Regularization methods can be used

Question 4

Logistic Regression is a ______________ algorithm, not a regression algorithm.

Answer 4

classification

Question 5

Default C value:

Answer 5

1

Question 6

T/F Using a higher value of C stresses the importance that each individual data point be classified correctly.

Answer 6

TRUE

Question 7

______________ and ____________ are a text-like used to count sparse data.

Answer 7

BerniuolliNB
MultinomialNB

Question 8

T/F In Naïve Bayes, Larger alpha decreases the complexity of the model but does not change the performance.

Answer 8

TRUE

Question 9

______________ is a high-dimensional dataset.

Answer 9

GaussianNB

Question 10

It is a family of classifiers that are quite similar to the linear models.

Answer 10

Naïve Bayes Classifier

Question 11

T/F Logistic Regression is also a regression algorithm.

Answer 11

FALSE

Question 12

A regression where the dependent variable is in an ordered state (i.e., ordinal). The dependent variable (y) specifies an order with two or more categories or levels.

Answer 12

Ordinal Logistic Regression

Question 13

Linear SVM is implemented in svm.LinearSVC.

SVC stands for

Answer 13

Support Vector Classifier

Question 14

It predicts the relationship between the independent and binary dependent variables.

Answer 14

Binary Logistic Regression

Question 15

For LogisticRegression and LinearSVC the trade-off parameter that determines the strength of the regularization is called ______.

Answer 15

C

Question 16

A regression type that has more than two possible outcomes.

Answer 16

Multinomial Logistic Regression

Question 17

It is a statistical technique used to develop predictive models with categorical dependent variables having dichotomous or binary outcomes

Answer 17

Binary Logistic Regression

Question 18

T/F The reason that Naïve Bayes are so efficient is that they learn parameters by looking at each feature individually and collect simple per-class statistics from each feature.

Answer 18

TRUE

Question 19

The two most common linear classification algorithms

Answer 19

Logistic Regression
Linear Support Vector Machines (Linear SVMs)

Question 20

In Naïve Bayes, generalization performance is ___________________.

Answer 20

slightly slower

Question 21

A categorical dependent variable has two or more discrete outcomes in a __________ regression type.

Answer 21

Multinomial

Multinomial Logistic Regression

Question 22

T/F Naïve Bayes works well with sparse high-dimensional datasets and is not parameter sensitive.

Answer 22

TRUE

Question 23

Higher values of C correspond to _________ regularization.

Answer 23

less

Question 24

In Naïve Bayes, training speed is ____________ than linear classifier.

Answer 24

faster

Question 25

T/F Using low values of C will cause the algorithms to try to adjust to the “majority” of data points.

Answer 25

TRUE