TOPIC 3 - BIG DATA, DATA ANALYTICS, AND MACHINE LEARNING

Question 1

How can data help businesses

Answer 1

Smarter and faster decisions
Accurate predictions
Sorting the signal from the noise
Efficient operations including real-time changes

Question 2

Data is often viewed as what by companies

Answer 2

An asset

Question 3

What is the cross industry standard process for data mining

Answer 3

An iterative process, that often involves going back-and-forth between stage

Question 4

In practice what should happen in cross industry standard process for data mining

Answer 4

Shortcuts from each stage back to the prior one

Question 5

What does CRISP-DM stand for

Answer 5

Cross industry standard process for data mining

Question 6

Whats the starting point and goal of CRISP-DM

Answer 6

Solve a business problem

Question 7

What should the business problem

Answer 7

Important and solvable

Question 8

How will the solution to the business problem be built

Answer 8

By using data as the raw material

Question 9

What do you need to understand

Answer 9

The strengths and weaknesses of the data

Question 10

Crisp-dm needs to weigh up what

Answer 10

Benefits and costs of aquiring

Question 11

What happens in the preparation stage

Answer 11

Clean the data
Decide on which variables you require

Question 12

What happens when you “clean” the data

Answer 12

Convert data to different types
Deal with missing values
Normalize or scale variables

Question 13

how do you decide on which variables you require

Answer 13

Can be guided by theory
In machine learning known as “feature engineering”

Question 14

What happens in the modelling and evaluation stages

Answer 14

May use various tools to help model the data
Need to evaluate our model rigorously
Important that your model is comprehensible

Question 15

why do we evaluate our model

Answer 15

Beware of correlations by chance, p-hacking and overfitting

Question 16

What happens in the deployment stage

Answer 16

Important to understand the benefits and risks of deployment
Continuous monitoring is often required

Question 17

How is continuous monitoring required

Answer 17

Such monitoring detects worsening or unexpected model performance
Allows timely remediation actions such as adding new variables or retraining your model

Question 18

Where does big data come from

Answer 18

Everywhere

Question 19

Specific examples of big data

Answer 19

Internet interactions
Text documents
images and videos

Question 20

Whats the widely used definition of big data

Answer 20

Big data is any set of data that is too large or too complex to be handled using conventional data-processing techniques

Question 21

Whats a synonym for big data

Answer 21

Alternative data

Question 22

What is the 4 v’s

Answer 22

Volume
Velocity
variety
varacity

Question 23

What is Volume in 4 Vs of big data

Answer 23

terabytes to exabytes of existing data to process

Question 24

What is Velocity in 4 Vs of big data

Answer 24

Streaming data, milliseconds to seconds to respond

Question 25

What is Variety in 4 Vs of big data

Answer 25

Structured, unstructured, text multimedia

Question 26

What is Varacity in 4 Vs of big data

Answer 26

Uncertainty due to data inconsistency and incompleteness

Question 27

How is data when talking about volume

Answer 27

data at rest

Question 28

How is data when talking about velocity

Answer 28

Data in motion

Question 29

How is data when talking about variety

Answer 29

Data in many forms

Question 30

How is data when talking about varacity

Answer 30

Data in doubt

Question 31

What are the additional Vs

Answer 31

Variability Value proposition

Question 32

What does the Vs of big data mean is not possible

Answer 32

To store and process all the data on a single computer

Question 33

What can big data help with

Answer 33

Process efficiency Finding new connections in data Improving predictions Needs to be analysed to have value

Question 34

Data analytics is the process of what

Answer 34

Discovering patterns and relationships in data

Question 35

What are the 4 types of analytics

Answer 35

Descriptive analytics Diagnostic analytics Predictive analytics Prescriptive analytics

Question 36

What is descriptive analytics and an example

Answer 36

What has happened -> describes something that happened -> 50% returns in a month

Question 37

What is diagnostic analytics

Answer 37

Why has something happened -> Describes the reason for the historical results -> Customers often return as not what they expected

Question 38

Predictive analytics and an example

Answer 38

What will happen if? -> determines what will happen by analyzing past data -> next quarter looks like a decline

Question 39

What is prescriptive analytics

Answer 39

What to do to make it happen -> Use info from other 3 to suggest a decision

Question 40

What is artificial intelligence

Answer 40

Computer models or systems that exhibit intelligent behavior like humans

Question 41

We currently have what

Answer 41

narrow ai systems

Question 42

What is Narrow AI

Answer 42

Specialize in specific tasks

Question 43

What is machine learning

Answer 43

Study of algorithms that: - improve their performance - at some task - with experience

Question 44

Example of machine learning

Answer 44

T: Playing chess P: Percent games won against an opponent E: Playing games against itself

Question 45

What are the 2 types of machine learning

Answer 45

Supervised Unsupervised

Question 46

What is supervised machine learning

Answer 46

have training data with desired outputs (labels) needs a stable environment Focus on prediction

Question 47

What is unsupervised machine learning

Answer 47

Only have training data without labels No feedback Focus on finding groups of similar items based on the data

Question 48

What are the 5 approaches of data analysis

Answer 48

Traditional econometrics Supervised learning Unsupervised Traditional programming machine learning programs

Question 49

What approaches have labeled data

Answer 49

Traditional econometrics Supervised learning

Question 50

What has unlabeled data

Answer 50

Unsupervised learning

Question 51

What method is traditional econometrics

Answer 51

Linear regression

Question 52

What methods are supervised learning and unsupervised learning

Answer 52

Supervised Machine learning Unsupervised machine learning

Question 53

Results in traditional econometrics

Answer 53

Explanatory model and statistical significance

Question 54

what is the results of supervised learning

Answer 54

Prediction model and prediction performance

Question 55

What is the results for unsupervised learning

Answer 55

Data structure model and data structure characteristics

Question 56

what is traditional programming

Answer 56

Write a program with explicit rules to follow

Question 57

What is machine learning programs

Answer 57

write a computer program to learn from examples

Question 58

Supervised machine learning uses data for what

Answer 58

To learn a hypothesis to predict

Question 59

Supervised machine learning uses what

Answer 59

classification models Regression models

Question 60

When do you use classification models vs Regression models

Answer 60

Class -> Target variable categorical Reg -> target variable cont

Question 61

Whats optimization in Supervised machine learning

Answer 61

How is the model trained on the data

Question 62

whats representation for Supervised machine learning

Answer 62

How is the data specified What is the form of the model

Question 63

whats evaluation in Supervised machine learning

Answer 63

How are we assessing if model is successful Whats the performance measure

Question 64

AI wins when info is what

Answer 64

More transparent and voluminous

Question 65

Humans win when institutional knowledge is what

Answer 65

Crucial

Question 66

Performance edge of Ai what over time

Answer 66

Declines over time when alternative data is found

Question 67

Combing ai and main produces what

Answer 67

The most accurate forecasts

Question 68

Applications of AI and ML in finance

Answer 68

asset management call centres credit and insurance

Question 69

When will larger training datasets improve prediction accuracy

Answer 69

if given X a human can confidently predict Y then yes

Question 70

ML techniques are valuable when:

Answer 70

Have lots of features and training examples Impact of features is highly nonlinear prediction is more important that inference

Question 71

Some ML approaches require alot of what

Answer 71

Computing power

Question 72

Whats one solution to ML needing high computing power

Answer 72

Cloud computing services