Introduction and Applications

Question 1

Why would data-driven prediction be useful?

Answer 1

Finding non-obvious patterns (Hurricane Frances)

Question 2

What is:

Data science?

Answer 2

It’s a set of fundamental principles that guide the extraction of knowledge from data

Question 3

What is:

Data science?

Answer 3

It’s a set of fundamental principles that guide the extraction of knowledge from data

Question 4

What is:

Data mining?

Answer 4

It is the automatic extraction of patterns from data (via tools/technologies that incorporate the principles)

Question 5

What is:

Big Data?

Answer 5

Big Data is data that is so large that traditional data processing systems are unable to deal with it (both storage and analysis component)

Question 6

What is:

Querying?

Question 7

What is:

OLAP?

Answer 7

Short for “On-Line Analytical Processing”, OLAP is an advanced uery and reporting technique where there is a multidimentional analysis of data input. But is does not automatically extract patterns.

Question 8

What is:

Business Intelligence?

Answer 8

Getting the right information to the right person at the right
time

Question 9

What is:

Data warehousing?

Answer 9

Data warehousing is the collection and coalescence of data from across an enterprise, often from multiple transaction-processing systems, each with its own database.

Question 10

What is:

An ‘instance’?

Answer 10

It is a vector of size [#amount of input variables or features]
There are [#observations in data set] data instances

Question 11

What is:

A feature?

Answer 11

Also called input variable
It is a vector of size [#number of data instances]
There are [#number of input variables] features

Question 12

What is:

Machine Learning?

Answer 12

It is improving the knowledge of a learning agent by providing data to it, it transcends data mining, since it uses robotics…

Question 13

What is:

Artificial Intelligence?

Answer 13

It is the automatic extraction of patterns from large amounts of data. E.g. the computer than can interact due to data (Big Data + Machine Learning = Artificial Intelligence)

Question 14

What is:

The difference between ‘data mining’ and ‘using data mining results’?

Answer 14

[1] Is the mining of historical data to produce a model that tries to predict a target variable (for supervised learning)
[2] The phase where the extracted model is applied to new data for which the class value is unknown

Question 15

What is:

CRISP-DM?

Answer 15

Short for ‘Cross Industry Standard Process for Data Mining’, CRISP-DM means a process for data mining consisting of (1) Business understanding, (2) Data understanding, (3) Data preparation, (4) Modeling, (5) Evaluation, and (6) Deployment
It is a process that should go through multiple iterations