quartz/content/notes/entity-relationship-diagrams.md
2022-04-07 01:50:11 +12:00

48 lines
1.1 KiB
Markdown

---
title: "entity-relationship-diagrams"
tags:
- info201
---
![](https://i.imgur.com/SIgTzZ5.png)
cardinality
identifying vs non identifying relationship
labels are important - but not always needed
associative entity => changes many to many relationship with additional relationship
## 1 subtypes
![](https://i.imgur.com/5sgPCxO.png)
![](https://i.imgur.com/Q0jMI3b.png)
uses:
- model mutual exclusivity
- better for modelling not for implementation
## 2 parallel relationship
![](https://i.imgur.com/UJXPI1l.png)
could model as separate relationships via staff subtypes
not very common
![](https://i.imgur.com/niEL1Y2.png)
also an example of recursive many-to-many relationships
## 3 recursive relationship
labels are critical
usually 1:M can be 1:1 or M:M
![](https://i.imgur.com/CaEgEkp.png)
## 4 dealing with data history
![](https://i.imgur.com/cohxggK.png)
could be many to many relationships:![](https://i.imgur.com/g4ynsh2.png)
so associative relationship: ![](https://i.imgur.com/NXxsJRl.png)
what do we require:
- for the current point in time
- an histroical record how ⇒ must be selecetive to not use up to much space