CS35 - First Quiz

Question 1

The mathematical discipline that studies the methods of collecting, analysing, and interpreting data.

Answer 1

Statistics

Question 2

Subset or subcollection of the population

Answer 2

Sample

Question 3

Specific collection of items of interest

Answer 3

Population

Question 4

Logic is built based on business rules

Answer 4

Traditional Rule-Based AI

Question 5

Logic is built by modelling and training data

Answer 5

Machine Learning

Question 6

Machine learning algorithm in which the training data includes both input and output

Answer 6

Supervised

Question 7

Training data consists of only input without any known output

Answer 7

Unsupervised

Question 8

The model predicts whether a record is an instance of a specific class or category

Answer 8

Binary Classification

Question 8

The label predicted by the model is a numeric value

Answer 8

Regression

Question 9

The model predicts whether a record is an instance of one of multiple classes or categories

Answer 9

Multiclass Classification

Question 10

Model identifies similarities between observations based on their features and groups them into discrete clusters

Answer 10

Clustering

Question 11

Inputs are called

Answer 11

Feature values

Question 12

Outputs are called

Answer 12

Label values

Question 13

Proportion of prediction that the model got right

Answer 13

Accuracy

Question 14

Proportion of predicted positive cases where the true label is actually positive

Answer 14

Precision

Question 15

Proportion of positive cases that the model identified correctly

Answer 15

Recall

Question 16

Overall metric combining recall and precision

Answer 16

F1 Score

Question 17

Formula of accuracy

Answer 17

(TN + TP)/(TN + FN + FP + TP)

Question 18

Formula of precision

Answer 18

TP / (TP + FP)

Question 19

Formula of recall

Answer 19

TP / (TP + FN)

Question 20

Formula of f1 score

Answer 20

(2 x Precision x Recall) / (Precision + Recall)

Question 21

Available data set is split into:

Answer 21

Training data and test data

Question 22

Typical training data is around:

Answer 22

70% to 80%

Question 23

A technique to balance uneven datasets by keeping all of the data in the minority class and decreasing the size of the majority class.

Answer 23

Undersampling

Question 24

It works by creating synthetic examples for the minority class rather than simply duplicating existing instances. These synthetic examples are generated by interpolating between existing instances of the minority class.

Answer 24

Synthetic Minority Oversampling Technique)

Question 25

Occurs when one class is significantly more frequent than the other.

Answer 25

Class Imbalance

Question 26

Used to address imbalance.

Answer 26

Undersampling the minority class