DATA2001 · Data Science, Big Data and Data Variety
Foundations and ER Modelling
Week 1 of University of Sydney DATA2001 sets the framing for the whole unit — the data/information/knowledge distinction, the V's of big data, and where data lives (warehouse vs lake vs lakehouse) — then teaches conceptual database design with the Entity-Relationship model (Peter Chen, 1976): entities, attribute kinds, keys, relationship degree and mapping cardinality. This material is tested in the early weekly quizzes and the Early Feedback Task, and ER/keys reasoning underpins every later chapter and the 50% final exam.
What this chapter covers
- 01Data vs information vs knowledge (the DIKW distinction) and the V's of big data — volume, velocity, variety, variability, veracity, visualization, value
- 02Storage architectures: data warehouse (structured, SQL-optimised) vs data lake (raw, native format) vs data lakehouse (hybrid)
- 03Conceptual database design: capturing information requirements graphically, independent of physical implementation
- 04Entity sets (rectangles) and attribute kinds — simple, composite, multi-valued (double ellipse), derived
- 05Keys: super key, candidate key (minimal), primary key (underlined; unique, no NULLs)
- 06Relationships (diamonds), degree (unary/recursive, binary, ternary), and the relationship key as the combined PKs
- 07Mapping cardinality (1:1, 1:N, N:1, M:N) and constraints — key (arrow), participation (thick line, total vs partial), min..max annotation
Reading cardinality, key and participation constraints from a sentence
- +1Mapping cardinality. Each tutorial class relates to at most one tutor, while a tutor relates to many classes — reading both directions, one tutor to many classes and each class to one tutor is a one-to-many (1:N) relationship from Tutor to TutorialClass (equivalently N:1 from TutorialClass to Tutor).
- +1Key constraint. Because each TutorialClass participates in at most one Leads instance (exactly one tutor), the key constraint is on the TutorialClass side — drawn as an arrow from the diamond to TutorialClass. Tutor has no key constraint (it can lead many classes).
- +1Participation. Rule (iii) says every class must have a tutor, so TutorialClass has total participation (thick line) — every class entity is in at least one Leads relationship. Rule (ii) says some tutors lead none, so Tutor has partial participation.
- +1min..max on TutorialClass. Each class relates to exactly one tutor — a minimum of 1 (total participation) and a maximum of 1 (key constraint) — so the annotation is 1..1. TutorialClass therefore has exactly-one participation, drawn as a thick arrow.
Key terms
- Entity set
- A collection of distinguishable objects of the same kind (students, courses, sensors) that share attributes; drawn as a rectangle in an ER diagram. An individual member is an entity.
- Multi-valued attribute
- An attribute that can hold several values for one entity (e.g. a book's authors or a person's phone numbers); drawn as a double ellipse, unlike a single-valued attribute's single ellipse. A derived attribute (e.g. age from date of birth) is drawn dashed.
- Super key vs candidate key
- A super key is any attribute set that uniquely identifies an entity; a candidate key is a minimal super key (no attribute can be removed while keeping uniqueness). The primary key is one chosen candidate key, shown underlined.
- Degree of a relationship
- The number of entity sets a relationship connects: unary/recursive (one set related to itself, e.g. Employee manages Employee), binary (two sets), or ternary (three sets).
- Mapping cardinality
- For a binary relationship, how many entities of each side can relate to one of the other: 1-to-1, 1-to-Many, Many-to-1, or Many-to-Many. Considered in both directions.
- Participation constraint
- Whether every entity of a set must take part in the relationship: total participation (at least one, drawn with a thick line) vs partial. Combined with a key constraint it becomes 'exactly one' (thick arrow, min..max = 1..1).
Foundations and ER Modelling FAQ
What is the difference between data, information and knowledge?
Data is the raw, context-free material — numbers, text, symbols. Information is data processed and placed in a context so it means something to a user. Knowledge is the ability to understand information, form a view and make decisions or predictions from it. DATA2001 opens with this DIKW ladder because the whole unit is about turning raw, varied data into information you can act on.
Why do we draw an ER diagram before writing any SQL?
Conceptual design captures the business's information requirements graphically and independently of how the data will be stored, so technical and non-technical stakeholders can agree on the model first. The ER diagram (entities, attributes, relationships, keys, cardinalities) is then mechanically mapped to a relational schema in Week 2. Getting the entities, keys and cardinalities right on paper prevents expensive schema mistakes later.
How do I choose the primary key?
List the candidate keys — the minimal attribute sets that uniquely identify each entity — then pick one as the primary key. Prefer a small, stable, never-NULL identifier (often a surrogate id). When no single attribute is unique, you may need a composite key: for example, a house number is not unique on its own, so (house_number, street_name) together form the key. The primary key is underlined in the diagram and disallows NULLs.
Can AI help me with ER modelling in DATA2001?
Yes, as a study aid. Sia can walk you through turning a set of business rules into entities, attributes, keys and cardinality/participation constraints, and check your arrows, thick lines and min..max annotations against the standard notation. Bring your own tutorial scenario and work it step by step. It explains the method and rehearses you; it does not complete graded assessment, and the University of Sydney academic-integrity policy applies.
Exam move
Make the ER notation muscle memory: rectangles for entity sets, ellipses for attributes (double for multi-valued, dashed for derived), diamonds for relationships, underline the primary key, arrow for a key constraint, thick line for total participation, and min..max on the edge. Practise by converting a paragraph of business rules into a diagram and back into prose — the weekly quiz and Early Feedback Task love the 'exactly one vs at least one vs at most one' distinctions. Learn the V's of big data well enough to give an example of each, and be able to contrast a data warehouse, lake and lakehouse in a sentence. This chapter is low on calculation but high on precise vocabulary, so quiz yourself on definitions. Ask Sia to generate fresh rule-to-diagram drills, and confirm any assessment specifics on Canvas.
Working through Foundations and ER Modelling in DATA2001? Sia is AskSia’s AI Data Science tutor — ask any DATA2001 Foundations and ER Modelling question and get a clear, step-by-step explanation grounded in how DATA2001 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
