Defining Data science

Question 1

What is data science

Answer 1

Art of processing data to find answers to your question

Analyzing data and trying to get answers

Question 2

What we can get from data science

Answer 2

1- We can be descriptive ( summarize characteristics of data set, no interpretation)
2- We can be exploratory ( explore patterns, trends and relationships within a dataset; we use to generate hypothesis for future investigation)
3- It can be predictive ( in order to predict an outcome)
4- It can be inferential ( derive conclusions about data set)
5- It can be “causal” not “casual” (if we change one factor will it lead to the change of other factors
6- About underlying mechanisms of the observed patterns

Question 3

Framework of doing data science:

Answer 3

1- Problem Identification
2- Data discovery (through screening , inventory, or aquisition)
3- Data ingestion and governance
4- Data wrangling
5- Fitness for use
6- Statistical modelling and analysis
7- Communication and dissemination
8- All that must have Ethics Review

Question 4

Factors affecting problem Identification

Answer 4

1- Theories and Hypothesis
2- Domain expertise
3- Domain knowledge

Question 5

Factors affecting data discovery

Answer 5

1- Potential data sources

2- Data integration

Question 6

Factors affecting data discovery

Answer 6

1- Potential data sources

2- Data integration

Question 7

Data collection types

Answer 7

1- Statistically designed ( surveys, experiments, remote sensing)
2- Adminstrative ( governmental agnecies, registered student data)
3- Oppurtunity ( from the internet - API)
4- Procedural ( Related to process and policies like change in insurance policy)

Question 8

Data Wrangling

Answer 8

Transforming raw data into appropriate form for analysis

Question 9

Data governance and ingestion

Answer 9

Data governance: establishment and adherence to rules regarding data access, dissemination and destruction

Question 10

Data ingestion

Answer 10

Bringing data into data management platforms

Question 11

Fitness for use assessment

Answer 11

Assessing the constraints on data by the statistical methods

Question 12

Data analysis type

Answer 12

1-Summarization
2-Visualization
3-Classification: predicting category for new data
4- Regression: Predicting quantitative value for new observation
5- Clustering finding unlabeled subgroups
6- Estimation: taking measurements for small numbers in a large group and making a good guess for the large group

Question 13

Variable types:

Quantitative and Qualitative

Answer 13

Quantitative : measurements are close in value and nature like pressure of wind
Qualitative: assumes values in finite set , also known as categories, discrete variables and factors

Question 14

Attribute:

Answer 14

Data field representing characteristic of data object

Question 15

Qualitative attribute- Nominal

Answer 15

Like patient ID , occupation

Question 16

Qualitative attribute - Binary

Answer 16

Can have two value like True or False

Question 17

Qualitative attribute - ordinal

Answer 17

Have meaningful order like pain level moderate severe minimal

Question 18

Quantitative - measurable quantity ;

Interval- scaled measured on an equal sized units

Answer 18

1- Do not have a zero-point

2- Mean, median, and mode are meaningful

Question 19

Quantitative - Ratio-scaled

Answer 19

ordered integer values with a zero point like weight

Question 20

Predicting output using input

Answer 20

Regression for quantitative outputs

Classification for qualitative output

Question 21

Data frame is like a table. they are rectangular objects containing data

Answer 21

True

Question 22

Every column of data frame is a vector - We must have same vector length over a data frame

Answer 22

True

Question 23

Vector vs list

Answer 23

Vector: all data same type
List: different data types: integer, character, logical, list