Data Science DEP summary

Data Science DEP
Lecture notes and some additional info from the suggested book chapter

Suggested book chapter: C.S. Jensen, T.B. Pedersen, C. Thomsen, "Fundamental
Concepts". Chapter 2 in "Multidimensional Databases and Data Warehousing".
2010. Morgan & Claypool. ISBN 978-1608455379.

, How to get insights from data
1. Formulate “questions to data”
2. Imagine visualizations/reports
3. Design star schema(s) for cube(s) by analyzing (1) and (2) for fact(s) and dimensions
4. Create (empty) database with schema
5. Fill database by transforming sources
6. Use: Analytics or (Predictive) modeling by connecting to the database

How to go from sources to senses: Transform (re-shape) sources into cubes in a database
(DMBS, safe space) and access (connect) through visualization environment and analytical
applications

Data science process
1. Sources
2. Prepare
3. Analyze
4. Use

Sources
1. Information systems
2. Sensors
3. Internet
4. Social media

Prepare
1. Search
2. Harvest
3. Combine
4. Transform
5. Clean

Analyze
1. Machine learning
2. Mining
3. Visualize

Use
1. Interpret
2. Deploy
3. Decide
4. Monitor

* Data preparation and cleaning is 80% of scientists time