data transformation

Question 1

What are the 6 ways of dealing with missing values?

Answer 1

1) ignore the tuple
2) fill in manually
3) use global constant
4) measure of central tendency
5) use mean/median for all samples belonging to same class as the tuple
6) use most probable value to fill in missing value

Question 2

What is a nominal feature?

Answer 2

Categorical variable relating to names where each value represents some kind of category in no meaningful order

Question 3

(T/F) Nominal data can be represented with numbers

Answer 3

True

Question 4

What is a binary feature?

Answer 4

0s and 1s, symmetric if both states are equally valuable and carry the same weight and asymmetric if outcomes are not equally important

Question 5

What is an ordinal feature?

Answer 5

A qualitative variable with possible values that have a meaningful order or ranking, but magnitude b/w successive values are not known

Question 6

What are examples of ordinal features?

Answer 6

drink size, grade, professional rank

Question 7

How can the central tendency of an ordinal feature be described?

Answer 7

by the mode or median

Question 8

How can numeric data be represented as a classifier?

Answer 8

as-is or normalized

Question 9

How can binary data be represented as a classifier?

Answer 9

0s and 1s

Question 10

How can ordinal data be represented as a classifier?

Answer 10

ordered numeric

Question 11

How can nominal data be represented as a classifier?

Answer 11

one-hot, numeric proxy

Question 12

What are the 6 strategies for data transformation?

Answer 12

1) Smoothing
2) Attribute clustering
3) Aggregation
4) Normalization
5) Discretization
6) Concept hierarchy generation

Question 13

What are the four different types of normalization techniques?

Answer 13

1) min-max
2) z-score
3) mean absolute deviation
4) decimal-scaling

Question 14

What is min-max normalization?

Answer 14

Performs linear transformation on original data while preserving original relationships

Question 15

What is z-score normalization?

Answer 15

(zero-mean), values for attribute A are normalized based on the mean and standard deviation of A

Question 16

What is the mean absolute deviation?

Answer 16

The absolute value of deviation from the mean, used bc it’s more robust to outliers than standard deviation

Question 17

What is decimal-scaling normalization?

Answer 17

Moves decimal point of values of attribute A where the number of decimal points depends on the maximum absolute value of A

Question 18

What are three methods of discretization?

Answer 18

1) Binning
2) Histogram analysis
3) Cluster, DT, and correlation analyses

Question 19

What is discretization?

Answer 19

Where raw values of numeric attributes are replaced by interval labels or conceptual labels. can also be a form of data reduction.

Question 20

What are three forms of feature engineering?

Answer 20

1) Summarization: replacing multiple instances with column value averages
2) Kernelization: Take existing features and explode into high-dimensional space, or passing a set of features through a function that produces higher-order features
3) Representation learning: let neural network learn feature space in the form of a fector

Question 21

Where does the target variable come from?

Answer 21

1) an existing feature that may be missing from some instances
2) a hand labeled feature
3) a feature value that will be known in the future