LU1- Recap

Question 1

What are the four types of Data Analytics?

Answer 1

Descriptive
Diagnostic
Predictive
Prescriptive

Question 2

Name 5 statistical techniques for Data Analysis

Answer 2

Linear regression
Classification
Resampling methods
Tree based methods
Unsupervised learning

Question 3

What is Linear Regression?

Answer 3

Linear Regression is the technique that is used to predict a target variable by providing the best linear relationship among the dependent and independent
variable where best fit indicates the sum of all the distances amidst the shape and
actual observations at each data point is as minimum as achievable.

Question 4

What is classification

Answer 4

Classification allocates specific
categories to a collection of data for making more spesific predictions and analysis.

Question 5

Name two types of classification techniques

Answer 5

Logistic Regression
Discriminant Analysis

Question 6

What is Logistic Regression?

Answer 6

A regression analysis technique to perform when the dependent variable is binary. It is a predictive analysis that is utilized for explaining data and the connection amongst one dependent variable and other nominal independent variables.

Question 7

Name two resampling techniques

Answer 7

Bootstrapping
Cross- Validation

Question 8

What is bootstrapping?

Answer 8

It operates through sampling with replacement from the actual
data and accounts the “not selected” data points as test samples.

Question 9

What is Cross-Validation?

Answer 9

This technique is used in order to validate the model performance, and
can be executed by dividing the training data into K parts. During cross validation execution, the K-1 part can be considered as training and the rest made out part acts
as a test set. Up to K times, the process is repeated and then the average of K scores is
accepted as performance estimation

Question 10

When does Undersampling take place?

Answer 10

When the majority of the class is copied

Question 11

When does oversampling take place?

Answer 11

When the minority of the class gets copied

Question 12

Name 3 Unsupervised learning algorithms

Answer 12

Principal component Analysis
K-Means Clustering
Hierarchical Clustering

Question 13

What is Principal component analysis?

Answer 13

recognising a linear-set of the mutually uncorrelated blend of features having maximum variance. Also, it helps in acquiring latent interaction among the
variables in an unsupervised framework.

Question 14

What is Machine Learning?

Answer 14

Machine Learning is the adoption of mathematical and or statistical models in order to get customized knowledge about data for making foresight.

Question 15

Name an unsupervised machine-learning technique

Answer 15

Clustering

Question 16

What are Latent variable models?

Answer 16

Latent variable models are commonly used for data preprocessing, such as reducing the number of features in a dataset (dimensionality reduction) or decomposing the dataset into multiple components.

Question 17

What is Clustering

Answer 17

A clustering problem is where you want to discover the inherent groupings in the
data, such as grouping customers by purchasing behavior.

Question 18

what type of learning( supervised/ unsupervised) makes use of clustering?

Answer 18

unsupervised

Question 18

what type of learning( supervised/ unsupervised) makes use of Classification?

Answer 18

Supervised

Question 19

what type of learning( supervised/ unsupervised) makes use of Regression?

Answer 19

Supervised

Question 20

What type of machine learning type makes use of labelled input and output data during the training phase of the machine learning lifecycle

Answer 20

Supervised learning

Question 21

To be able to classify new and unseen datasets and predict outcomes, what does a supervised learning model need to learn

Answer 21

relationship between input and output data

Question 22

What machine learning is where we have input variables (X) and an output variable (Y)

Answer 22

Supervised learning

Question 23

why do we call it supervised learning?

Answer 23

Because part of the approach requires human oversight

Question 24

what is classification?

Answer 24

Classification is used when the output variable is categorical

Question 25

Give an example of categorical data

Answer 25

yes/no or male/female or true/false

Question 26

what is regression

Answer 26

regression is used when the output variable is a real or continuous value

Question 27

give an example of regression variables

Answer 27

salary based on work experience or weight based on height

Question 28

give an example of an algorithm used for regression problem

Answer 28

Linear regression or support vector regression or regression tree

Question 29

give an example of a classification problem

Answer 29

The machine needs to understand the difference between stuff (apple, banana& cherry)

Question 30

what type of learning does not make use of output variables

Answer 30

unsupervised

Question 31

Unsupervised learning makes use of output variables (true or false)

Answer 31

False

Question 32

what do you call unsupervised learning output

Answer 32

pseudo output

Question 33

what is anomaly detection

Answer 33

It is when machine learning automatically detects unusual data points in a dataset

Question 34

what is association mining

Answer 34

Identifies sets of items that frequently occur together in your dataset

Question 35

what is latent variable models

Answer 35

Commonly used for data preprocessing such as reducing the number of features in a dataset

Question 36

Give a real world example of unsupervised learning

Answer 36

computer vision - object recognition
medical imaging
anomaly detection
customer personas- habits
recommendation engines

Question 37

what is the difference between unsupervised vs supervised

Answer 37

unsupervised - no output data is given
data is not labeled
computationally complex
less accurate & trustworthy
number of classes is not known

Question 38

what is accuracy in regards to supervised learning

Answer 38

the ability of a model to make correct predictions

Question 39

what is interpretability in regards to supervised learning

Answer 39

what degree the model allows for human understanding

Question 40

give an example of an interpretable model

Answer 40

linear regression
Random forest

Question 41

Give an example of a Non interpretable model

Answer 41

SVM(Support vector machine)
LSTM(Long short term memory)
Deep learning(DL)

Question 42

What is K-means clustering

Answer 42

A method of vector quantization, aims to partition n observations into k clusters in which each observation belongs to the cluster with the nearest mean

Question 43

what type of learning category does K- means clustering fall into

Answer 43

Unsupervised

Question 44

why use k- means clustering

Answer 44

K-means is used to classify unlabelled data by grouping them by features rather than categories. the goal is to split the data into k different clusters and report the location of the centre of mass for each cluster

Question 45

what does the K represent in K means cluster

Answer 45

the K represents the number of groups or categories created.

Question 46

what is hierarchical clustering

Answer 46

algorithm that creates clusters that have predominant ordering from top to bottom

Question 47

what does hierarchical clustering do?

Answer 47

Hierarchical clustering separates data into groups based on some measure of similarity

Question 48

what is Agglomerative Hierarchical Clustering

Answer 48

(“bottom-up”) clustering starts with each
observation being its own cluster. They merge into subgroups as we move up the tree.

Question 49

what is divisive clustering

Answer 49

(“top-down”) clustering starts with one cluster of
all observations. The cluster is split into subgroups as we
move down the tree.