1.1.1 Introduction to Data Science - Data Analytic Thinking

Question 1

List examples of data mining in finance, marketing, and customer relationship management.

Answer 1

Finance: credit scoring, trading, fraud detection.

Marketing: targeted marketing, online advertising, cross selling.

CRM: reduce attrition by analyzing customer behavior

Question 2

Contrast data science with data mining.

Answer 2

Data science is a set of principles that suggests how to extract knowledge from data.

Data mining is using technology to extract knowledge from data by incorporating data science principles.

Data science if of often viewed more broadly than data mining

Question 3

Describe the two types of decisions that can benefit from data-driven decision making.

Answer 3

(1) Decisions for which “discoveries” need to be made within the data, and (2) decisions that repeat, at a massive scale, so decision-making can benefit from small increases in decision making accuracy.

Question 4

Describe the reason for the early adoption of automated decision making by finance and telecommunications industries.

Answer 4

These industries we’re the first to build large data networks and implementation of massive-scale computing which allowed the aggregation and analyzing of data.

Question 5

Contrast data science with data processing.

Answer 5

Data science needs access to data, which data processing technologies may facilitate, but are not data science technologies per se.

Question 6

Describe the usage of big data.

Answer 6

Often used for data processing in support of data mining techniques.

Question 7

Explain why both appropriate data and data scientists are required to extract useful knowledge from data.

Answer 7

Data science teams can yield little value w/o appropriate data; right data often cannot substantially improve decisions w/o the right data science talent.

Question 8

Explain why it is necessary to understand data science even if someone is not going to use data science directly.

Answer 8

It helps you spot obvious flaws and unrealistic assumptions in data mining projects

Question 9

List and describe the four fundamental concepts of data science.

Answer 9

1. The Cross Industry Standard Process for Data Mining (CRISP-DM)
2. From a large mass of data, technology can be used to find informative descriptive attributies of entities of interest
3. If you look to hard at a set of data, you will find something but it might not generalize beyond that data (overfitting)
4. Formulating data mining solutions and evaluating the resulta involves thinking carefully about the context in which they will be used.