Week 5: Data Exploring + Pre-Processing

Question 1

Mean

Answer 1

Average of all numbers

Question 2

Median

Answer 2

Middle number in a sequence

Question 3

Mode

Answer 3

Number that occurs most often within a set

Question 4

Range

Answer 4

Difference between highest and lowest values

Question 5

Standard Derivation is a measure used to

Answer 5

quantify the amount of variation of data values

Question 6

Histogram (2 points)

• is similiar to
• gives a rough sense of

Answer 6

• a bar chart but groups numbers into ranges (bins)
• density

Question 7

Name the distribution

Answer 7

normal

Question 8

Name the distribution

Answer 8

right skewed (where tail goes)

Question 9

Name type of distribution

Answer 9

Multimodal

Question 10

Draw

• positive linear association
• negative linear association
• non-linear associaition
• no association

Answer 10

Question 11

Scatter plots show…

Answer 11

how much one variable is affected by another

Question 12

Correlations show

Answer 12

how strongly pairs of variables are related

Question 13

What is the measure of correlation?

Answer 13

correlation coefficient r

1 is perfect

0 is no correlation

-1 is perfectly negative correlation

Question 14

An outlier is

Answer 14

an observation that lies an abnormal distance from other values in a random sample

Question 15

How do you identify outliers? (BPRD)

Answer 15

• box plot
• probablitity plot
• dions test
• rosners test

Question 16

Handling missing values on small scale < 5%

Answer 16

Drop or omit

Question 17

Handling missing values on larger scale methods (MKFS)

Answer 17

• Mean
• K-nearest neighbors
• fuzzy K-means
• singular value decomposition

Question 18

3 types of invalid data

Answer 18

missing data values

invalid values that suggest true values

invalid values that provide no information regarding true values

Question 19

What is scaling?

Answer 19

scaling features to lie between a given minium and maximum value

Question 20

Transformation is…

Answer 20

converting data from one format or structure into another format or structure

Question 21

Feature selection is…

Answer 21

the process of selecting a subset of relevant features for use in model construction

Question 22

4 Reasons for using feature selection (REIR)

Answer 22

reduces the complexity of a model

enables the machine learning algorithm to train faster

improve the accuracy if the right subset is chosen

reduces overfitting

Question 23

5 methods for dimensionality reduction

Answer 23

1. Decision Tree
2. Random forest
3. high correlation
4. factor analysis
5. principal component analysis

Question 24

Dimensionality reduction…

Answer 24

creates new combination of attributes