Data Preprocessing

Question 1

8 Step of Data Preprocessing

Answer 1

Data cleaning
Sampling
Aggregation
Discretization and Binarization
Feature Transformation & Scaling
Dimensionality Reduction
Feature subset selection
Feature Creation

Question 2

5 Type of Error Data

Answer 2

Missing data
Structural errors - Typographical errors and other inconsistencies.
Duplicate data
Irrelevant data
Outliers

Question 3

Sampling

Answer 3

The process of selecting a representative subset of individuals, items, or events from a larger population, in order to estimate or infer information about the population as a whole.

Question 4

2 Type of Sampling

Answer 4

Simple Random Sampling
Stratified Sampling

Question 5

4 Motivation of Aggregation

Answer 5

A change of scope or scale
Data reduction
Noise reduction
Computation

Question 6

1 Disadvantage of Aggregation

Answer 6

Potential loss of interesting details.

Question 7

3 Type of Unsupervised Discretization

Answer 7

Equal Interval Width
Equal Frequency
K-means - Divide into discrete group

Question 8

1 Type of Supervised Discretization

Answer 8

Decision Tree

Question 9

Feature Transformation & Scaling

Answer 9

A function that maps the entire set of values of a given attribute to a new set of replacement values.

Question 10

2 Algorithm sensitive to feature scaling

Answer 10

Gradient Descent Based Algorithms
Distance-Based Algorithm

Question 11

1 Algorithm insensitive to the scale of the features

Answer 11

Tree-Based Algorithms

Question 12

3 Type of Feature Transformation

Answer 12

|x|Transformation - Symmetric variable around zero.
1/x Transformation - Upper end of distribution to lower end.
Log Transform - Convert a skewed distribution to a normal distribution / less-skewed distribution.

Question 13

4 Type of Feature Scaling

Answer 13

Max Abs Scaling (Between -1 and +1)
Min-Max Scaling (Between 0 and +1)
Standardisation (Z-Score Normalisation) (Between -3 and +3)
Normal distribution / Gaussian distribution (Bell-shaped curve)

Question 14

Normal distribution

Answer 14

A statistical concept that describes a probability distribution of a random variable.

Question 15

Dimensionality Reduction

Answer 15

The process of reducing the number of input features in a dataset while preserving the key information or patterns.

Question 16

Principal Components / Latent Variables

Answer 16

Features that are combinations of the original features

Question 17

Principal Component Analysis (PCA)

Answer 17

A technique to transform a high-dimensional dataset into a lower dimensional space, while retaining as much of the information in the original data as possible.

Question 18

5 Motivations of Dimensionality Reduction

Answer 18

Reduced noise
Enhanced interpretability
To visualise the data and gain insights
To speed up a subsequent training algorithm
To save space (compression)

Question 19

4 Drawbacks of Dimensionality Reduction

Answer 19

Some information is lost
Transformed features are often hard to interpret
Can be computationally intensive
Adds some complexity to pipelines

Question 20

3 Approach of Feature Subset Selection

Answer 20

Embedded approaches - Occurs naturally as part of the data mining algorithm, e.g. Decision tree

Filter approaches - Selected before running the data mining algorithm; Low correlation

Wrapper approaches - Find the best subset of attributes
Forward selection - Take in one by one
Backward selection - Take out one by one
Bi-directional elimination (Stepwise Selection)