Chapter 9: Database Systems

Question 1

Databases

• Database:
• Database schema:
• Database management system (DBMS):
• Database model:

Answer 1

• Database: organized collection of data (Managed by DBMS)
• Database schema: description of the structure of the database
• Database management system (DBMS) - software that manages creation, update, querying and administration of databases.
• Database model: conceptual view (abstraction) of the database

Question 2

• Database systems

Steps

Answer 2

Step 1: User
Step 2: Application software
Step 3: Database management system
Step 4: Actual database

Question 3

The relational model

Answer 3

• Data is conceptually stored in table representing relations
• Tables have columns representing attributes and rows representing instances (records/tuples)
• The relational model for database management is an approach to managing data using a structure and language consistent with first order predicate logic.

Question 4

• Relational operations

Answer 4

• Select: Choose rows
• Project: Choose columns
• Join: Assemble information from two or more tables (relations)

Question 5

• SQL (Structured query language)
• What is a SQL?
• Difference between SQL and DBMS

Answer 5

• What is a SQL?
  SQL is used to communicate with a database and it’s the standard language for relational database management.
• Difference between SQL and DBMS
  We use the DBMS system along with SQL for querying data in the database. In SQL we can see the data we want by asking a query, while DBMS optimizes the query and shows us the data.

Question 6

• Operations to manipulate tables (relations)

Answer 6

INSERT
DELETE
UPDATE
SELECT

Question 7

• Object-oriented databases

Answer 7

• Object oriented database is constructed by applying the object-oriented paradigm.
• Each entity is stored as a persistent object, which could contain methods
• DBMS maintains inter-object links/references
• Matches the paradigm of object-oriented programming

Question 8

• Maintaining database integrity

Answer 8

• Transaction: sequence of operations, which should all together either succeed or fail
• Commit: the point at which a transaction is complete, confirmed and effectuated in the database
• Roll-back: undo the operations of a transaction (instead of committing them) because of a transaction conflict.

Question 9

• Data mining

Data warehouse

Strategies

Answer 9

• Data mining: deals with discovering patterns in collections of data
• Data warehouse: static database, which can be mined
• Strategies:
  
  • Class description
  • Class discrimination
  • Cluster analysis
  • Association analysis (Amazon)
  • Outlier analysis
  • Sequential analysis