DDMA Level II
Warehousing and Data Mining flashcards
Read the question, say the answer to yourself, then turn the card to check. The first 10 of 47 cards are free; the rest come with the subject.
Card 1 of 47
Jump to a card
Complete Flashcard Reference
Review every card in this deck. Open any card to reveal its answer. The first 10 are free; the other 37 come with the subject.
What is a data warehouse?
A central repository of integrated data drawn from one or more source systems. It is structured and stored to support analysis and reporting, not day-to-day transaction processing.
What are the four basic OLAP operations?
Roll-up: aggregates data to a higher, less detailed level, e.g. daily to monthly sales. Drill-down: the reverse, to a more detailed level, e.g. monthly to daily. Slice and dice: selects a subset of the data (slice) and views it from different angles (dice). Pivot (rotate): reorients the dimensions to view the data from a different perspective.
How does a snowflake schema differ from a star schema?
Star: a central fact table connected directly to denormalised dimension tables, simple and fast to query. Snowflake: extends the star by normalising the dimension tables into sub-dimensions, saving storage at the cost of more complex queries.
What does ETL stand for, and where does it sit in the data warehousing process?
Extraction, transformation and loading. It moves data from the operational source systems into the warehouse, from where it is queried and analysed.
What are the three types of data mart?
Dependent: built from data already in an existing central data warehouse. Independent: built directly from source systems, with no central warehouse behind it. Hybrid: draws from a central warehouse and from additional independent sources.
What is metadata, and why does it matter in a data warehouse?
Data about data: it describes the structure, source and meaning of the warehouse's content. It makes the data discoverable and trustworthy, telling a user or tool what a field means, where it came from and how current it is.
What is the difference between classification and prediction in data mining?
Classification assigns data to predefined categories, e.g. spam or not spam. Prediction estimates a future or unknown numeric value, e.g. next month's sales. Both use similar techniques; the output type, category versus number, is the deciding difference.
What does association analysis find, and what is the classic example?
Items that frequently occur together. The classic example is market-basket analysis, the retail application of data mining.
What are the five groups of issues in data mining?
Security and social: privacy risks and misuse of discovered patterns about individuals. User interface: making results interpretable and usable by non-technical decision makers. Mining methodology: choosing techniques that fit the data and question, avoiding spurious patterns. Performance: mining large datasets can be computationally expensive. Data source: mining is only as good as the underlying data's quality and completeness.
In a scenario, which technique is usually matched to retail, telecoms and security?
Retail: association analysis (market-basket). Telecoms: classification or prediction (customer churn). Security: outlier analysis (intrusion detection).
What are the four characteristics of a data warehouse?
This answer comes with Warehousing and Data Mining. Buy the subject
What does it mean that a data warehouse is non-volatile?
This answer comes with Warehousing and Data Mining. Buy the subject
What does it mean that a data warehouse is time-variant?
This answer comes with Warehousing and Data Mining. Buy the subject
What are the four components of a data warehouse, and what does each do?
This answer comes with Warehousing and Data Mining. Buy the subject
What are the layers of the basic data warehouse architecture?
This answer comes with Warehousing and Data Mining. Buy the subject
How does an OLTP system differ from a data warehouse?
This answer comes with Warehousing and Data Mining. Buy the subject
Who uses a data warehouse, and for what?
This answer comes with Warehousing and Data Mining. Buy the subject
What is Online Analytical Processing (OLAP)?
This answer comes with Warehousing and Data Mining. Buy the subject
How are roll-up and drill-down related?
This answer comes with Warehousing and Data Mining. Buy the subject
What do slice and dice and pivot have in common, as OLAP operations?
This answer comes with Warehousing and Data Mining. Buy the subject
What is a fact constellation schema?
This answer comes with Warehousing and Data Mining. Buy the subject
What are the advantages and disadvantages of data warehousing?
This answer comes with Warehousing and Data Mining. Buy the subject
What is a data mart?
This answer comes with Warehousing and Data Mining. Buy the subject
What are the three types of metadata?
This answer comes with Warehousing and Data Mining. Buy the subject
Why do the Metadata Interchange Initiative and its standard framework exist?
This answer comes with Warehousing and Data Mining. Buy the subject
What is a metadata repository?
This answer comes with Warehousing and Data Mining. Buy the subject
Why would an organisation create a data mart?
This answer comes with Warehousing and Data Mining. Buy the subject
What are the steps in implementing a data mart?
This answer comes with Warehousing and Data Mining. Buy the subject
What are the steps to implement a data warehouse?
This answer comes with Warehousing and Data Mining. Buy the subject
What are the three ways of coping with the business risks of a data warehouse implementation?
This answer comes with Warehousing and Data Mining. Buy the subject
Which risk strategy is the usual default for a first-time warehouse, and why?
This answer comes with Warehousing and Data Mining. Buy the subject
Name three data warehouse tools.
This answer comes with Warehousing and Data Mining. Buy the subject
What are the best practices for implementing a data warehouse?
This answer comes with Warehousing and Data Mining. Buy the subject
What is data mining?
This answer comes with Warehousing and Data Mining. Buy the subject
Name the data mining tools covered for this subject.
This answer comes with Warehousing and Data Mining. Buy the subject
What are the five data mining techniques, and what does each do?
This answer comes with Warehousing and Data Mining. Buy the subject
How do decision trees and neural networks differ as mining techniques?
This answer comes with Warehousing and Data Mining. Buy the subject
What criteria guide the choice of data mining software?
This answer comes with Warehousing and Data Mining. Buy the subject
What is the difference between characterization and discrimination?
This answer comes with Warehousing and Data Mining. Buy the subject
How does clustering differ from classification?
This answer comes with Warehousing and Data Mining. Buy the subject
What do outlier analysis and evolution and deviation analysis study?
This answer comes with Warehousing and Data Mining. Buy the subject
How is data mining used in financial analysis and in telecommunications?
This answer comes with Warehousing and Data Mining. Buy the subject
How is data mining used in retail and in intrusion detection?
This answer comes with Warehousing and Data Mining. Buy the subject
How is data mining used in higher education, the energy industry and manufacturing engineering?
This answer comes with Warehousing and Data Mining. Buy the subject
What do spatial data mining and biological data analysis look for?
This answer comes with Warehousing and Data Mining. Buy the subject
How is data mining used in criminal investigation and counter-terrorism?
This answer comes with Warehousing and Data Mining. Buy the subject
What are the technology trends in data mining?
This answer comes with Warehousing and Data Mining. Buy the subject
How to revise with these cards
- Answer out loud or on paper before you turn the card: recalling is what fixes it.
- Go through the deck in short sessions of ten to fifteen minutes, every day.
- Shuffle once you know the order, so you are not answering from memory of the sequence.
- When a card catches you out, read the chapter it links to, then try the topic in a past paper.