Notes on Entity Relationship Modelling and Diagrams
Database Creation Overview
- Creating a table is straightforward, and so is creating multiple tables.
- The complexity arises when deciding what data belongs in each table.
- Entity Relationship Models (ERM) and Entity Relationship Diagrams (ERD) help design the structure of database tables.
Entity Relationship Model & Diagram
Entity Relationship Model (ERM)
- A logical representation of data needed by an organization.
- Utilizes entities to symbolize people, objects, events, etc.
- Identifies relationships between various entities.
- Based on the business rules of the organization.
Entity Relationship Diagram (ERD)
- A graphical representation of the ERM.
- ER diagrams do not store data; they serve as a blueprint for database design.
ERDs vs Access Relationship Diagrams
- An ERD uses natural keys (real-world identifiers) without surrogate keys.
- An Access Relationship Diagram visualizes existing database tables with surrogate keys if used.
- ERDs are created prior to database construction; Access diagrams arise post-construction.
E-R Diagrams Components
Entities
- Main components in the E-R Diagram include entities representing objects like "Subject" and "Lecturer".
Attributes
- Attributes describe the properties of entities. For example, for Subject and Lecturer:
- Subject Attributes: SubjectCode, Title, CreditPoints.
- Lecturer Attributes: LecId, LecName, Age.
Relationships
- Relationships depict connections between entities. For example, there is a clear relationship between 'Subject' and 'Lecturer'.
E-R Diagrams & Sample Data
- When designing an ER Diagram, consider the actual data:
- Subject Data:
- INF10002: Database, Analysis & Design, Credits: 12.5
- INF10003: Intro to Business Info Systems, Credits: 12.5
- Lecturer Data:
- LecId: 207, Name: John Smith, Age: 37
- LecId: 119, Name: Jane Pitt, Age: 26
E-R Diagrams & Business Rules
- Business Rules examples:
- Students must enroll in one course at a time.
- A Student can enroll in multiple subjects.
- A subject can have only one convenor.
- An employee must have one tax file number.
- Some rules are common across organizations, while others are unique.
- Business rules should be documented as they guide data requirements.
Discovering Business Rules
- Challenges in discovering business rules include:
- No individual knows all rules.
- Individuals may be unaware of some rules or not disclose them due to fear or distrust.
- Lack of documentation can hinder rule identification.
- Business needs and rules evolve over time.
Cardinality Constraints
- Cardinality denoted how many instances of one entity relate to another.
- Examples for lecturers and subjects:
- One lecturer convenes one subject (drawn as ONE).
- One subject can have many lecturers (drawn as MANY or in a crow's foot).
Determining Cardinality
- Convened By Relationship:
- LEFT end: For a specific subject, only one lecturer can convene it (ONE).
- RIGHT end: A lecturer can convene multiple subjects (MANY).
- Read diagrams correctly to frame sentences:
- "One Subject is convened by ONE Lecturer".
- "One Lecturer convenes MANY Subjects".
Summary of Cardinality Constraints
- A single Subject is convened by ONE Lecturer, and ONE Lecturer can convene MANY Subjects.
- Initiate sentences with "ONE" to maintain clarity.
Entities and Instances
- An entity instance indicates a specific set of attribute values.
- Example: One instance could be LecId: 207, LecName: John Smith, Age: 37.
E-R Diagrams – Identifiers
- Identifiers uniquely identify an entity instance (similar to a Primary Key).
- Rules for identifiers include:
- Every entity must possess an identifier.
- Unique identifiers for entity instances – no duplicates.
- Identifiers must not be empty or null.
- Identifiers are underlined in the ERD.