ENTITY-RELATIONSHIP MODEL. CS 564- Spring 2018

Size: px

Start display at page:

Download "ENTITY-RELATIONSHIP MODEL. CS 564- Spring 2018"

Lillian Houston
6 years ago
Views:

1 ENTITY-RELATIONSHIP MODEL CS 564- Spring 2018

2 WHAT IS THIS LECTURE ABOUT E/R Model: entity sets, attribute relation: binary, multi-way relationship roles, attributes on relationships subclasses (ISA) weak entity sets constraints design principles E/R to Relational Model CS 564 [Spring 2018] - Paris Koutris 2

3 HOW TO BUILD A DB APPLICATION CS 564 [Spring 2018] - Paris Koutris 3

4 HOW TO BUILD A DB APPLICATION Pick an application Figure out what to model (ER model) Output: ER diagram Transform the ER diagram to a relational schema Refine the relational schema (normalization) Now ready to implement the schema and load the data! CS 564 [Spring 2018] - Paris Koutris 4

5 RUNNING EXAMPLE We want to store information about: companies and employees Each company has a name, an address, Each company has a list of employees products manufactured by these companies Each product has a name, a description, CS 564 [Spring 2018] - Paris Koutris 5

6 E/R MODEL Gives us a visual language to specify what information the DB must hold what are the relationships among components of that information Proposed by Peter Chen in 1976 What we will cover: 1. basic stuff: entities, attributes, relationships 2. constraints 3. weak entity sets 4. design principles CS 564 [Spring 2018] - Paris Koutris 6

7 ENTITIES & ATTRIBUTES Entity an object distinguishable from other object Entity set a collection of similar entities represented by rectangles described using a set of attributes Attribute represented by ovals attached to an entity set Product name CS 564 [Spring 2018] - Paris Koutris 7

8 ENTITIES & ATTRIBUTES price name category name stockprice Product Company Entities are not explicitly represented in E/R diagrams! Person name address ssn CS 564 [Spring 2018] - Paris Koutris 8

9 RELATIONS A mathematical definition: if A, B are sets, then a relation R is a subset of A x B Example A = {1, 2, 3}, B = {a, b, c, d} R = {(1, a), (1, c), (2, c), (3, b)} A a b c d B CS 564 [Spring 2018] - Paris Koutris 9

10 RELATIONSHIPS price name category name stockprice Product makes Company makes is a subset of Product x Company Person name address ssn CS 564 [Spring 2018] - Paris Koutris 10

11 RELATIONSHIPS price name category name stockprice Product makes Company buys Person employs name address ssn CS 564 [Spring 2018] - Paris Koutris 11

12 MULTIPLICITY OF RELATIONSHIPS one-one A a b c d B many-one A a b c d B many-many A a b c d B CS 564 [Spring 2018] - Paris Koutris 12

13 MULTIPLICITY OF RELATIONSHIPS price name category name stockprice Product makes Company buys Person employs name address ssn CS 564 [Spring 2018] - Paris Koutris 13

14 NOTATION DIFFERENCE We use: The cow book uses (page 33): You should use the notation in the slides! CS 564 [Spring 2018] - Paris Koutris 14

15 MULTI-WAY RELATIONSHIPS How do we model a purchase relation between buyers, products and stores? Product purchase Store Person We can still model this as a mathematical set (ternary relation) CS 564 [Spring 2018] - Paris Koutris 15

16 ARROWS IN MULTI-WAY RELATIONSHIPS What does the arrow mean here? Product purchase Store Person A given person can purchase a given product from at most one store! CS 564 [Spring 2018] - Paris Koutris 16

17 ARROWS IN MULTI-WAY RELATIONSHIPS What about here? Product purchase Store Person A given person can purchase a given product from at most one store AND a given store sells to a given person at most one product CS 564 [Spring 2018] - Paris Koutris 17

18 ARROWS IN MULTI-WAY RELATIONSHIPS How can we say that a given person buys from at most one store? Product purchase Store Person Not possible, we can only approximate! CS 564 [Spring 2018] - Paris Koutris 18

19 ROLES IN RELATIONSHIPS What if we need an entity set twice in a relationship? Product purchase Store Person salesperson Person buyer CS 564 [Spring 2018] - Paris Koutris 19

20 ROLES IN RELATIONSHIPS Label the edges to indicate the roles Collapse the two entity sets into one Product purchase Store salesperson buyer Person CS 564 [Spring 2018] - Paris Koutris 20

21 ATTRIBUTES IN RELATIONSHIPS date Product purchase Store Person Relationships have attributes as well! CS 564 [Spring 2018] - Paris Koutris 21

22 MULTI-WAY TO BINARY RELATIONSHIPS Product purchase Store productof Product Person Purchase storeof Store Why do we need the arrows? buyerof Person CS 564 [Spring 2018] - Paris Koutris 22

23 RELATIONSHIPS: RECAP Modeled as a mathematical set Binary and multi-way relationships Converting a multi-way one into many binary ones Constraints on the degree of the relationship many-one, one-one, many-many limitations of arrows Attributes of relationships not necessary, but useful! CS 564 [Spring 2018] - Paris Koutris 23

24 E/R: ADDITIONAL FEATURES CS 564 [Spring 2018] - Paris Koutris 24

25 SUBCLASSES subclass = specialized case = fewer entities = more properties Example: Products Software products Educational products CS 564 [Spring 2018] - Paris Koutris 25

26 SUBCLASSES price name category Product Attributes are inherited by the subclasses! version ISA age group Software Product Educational Product CS 564 [Spring 2018] - Paris Koutris 26

27 CONSTRAINTS constraint := an assertion about the database that must be true at all times part of the database schema central in database design When creating the ER diagram, you need to find as many constraints as possible! CS 564 [Spring 2018] - Paris Koutris 27

28 TYPES OF CONSTRAINTS keys: SSN uniquely identifies a person single-value: a person can have only one father referential integrity: if you work for a company, it must exist in the database domain: age is between 0 and 150 other: e.g. at most 80 students enroll in a class CS 564 [Spring 2018] - Paris Koutris 28

29 WHY DO WE NEED CONSTRAINTS? Give more semantics to the data help us better understand it Prevent wrong data entry Allow us to refer to entities (e.g. using keys) Enable efficient storage and data lookup CS 564 [Spring 2018] - Paris Koutris 29

30 KEY CONSTRAINTS price name category name stockprice Product makes Company buys Person employs name address ssn CS 564 [Spring 2018] - Paris Koutris 30

31 KEY CONSTRAINTS Every entity set must have a key! A key can consist of more than one attribute There can be more than one key for an entity set one key will be designated as primary key No formal way to specify multiple keys in an ER diagram CS 564 [Spring 2018] - Paris Koutris 31

32 SINGLE-VALUE CONSTRAINTS An entity may have at most one value for a given attribute or relationship an attribute of an entity set has a single value a many-one relation implies a single value constraint CS 564 [Spring 2018] - Paris Koutris 32

33 REFERENTIAL INTEGRITY CONSTRAINT A relationship has one value and the value must exist Product makes Company CS 564 [Spring 2018] - Paris Koutris 33

34 WEAK ENTITY SETS Entity sets are weak when their key attributes come from other classes to which they are related Team affiliation University number sport name entities of an entity set need help to identify them uniquely! CS 564 [Spring 2018] - Paris Koutris 34

35 E/R: DESIGN PRINCIPLES CS 564 [Spring 2018] - Paris Koutris 35

36 1. BE FAITHFUL TO THE APP! What is wrong here? Product purchase Person Country president Person CS 564 [Spring 2018] - Paris Koutris 36

37 2. AVOID REDUNDANCY! What is wrong here? date Product purchase Store personname personaddress CS 564 [Spring 2018] - Paris Koutris 37

38 2. AVOID REDUNDANCY! Redundancy occurs when we say the same thing in two different ways Redundancy wastes space and encourages inconsistency The two instances of the same fact may become inconsistent if we change one and forget to change the other CS 564 [Spring 2018] - Paris Koutris 38

39 2. AVOID REDUNDANCY! What is wrong here? address Beer manfby Manufacturer manufacturer name CS 564 [Spring 2018] - Paris Koutris 39

40 3. KEEP IT SIMPLE! What is wrong here? Date date Product purchase Store Person CS 564 [Spring 2018] - Paris Koutris 40

41 4. ATTRIBUTES OVER ENTITY SETS An entity set should satisfy at least one of the following conditions it is more than the name of something; it has at least one non-key attribute it is the many in a many-one or many-many relationship Beer manfby Manufacturer name What is wrong here? name CS 564 [Spring 2018] - Paris Koutris 41

42 5. DON T OVERUSE WEAK ENTITY SETS Beginner database designers often doubt that anything could be a key by itself They make all entity sets weak, supported by all other entity sets to which they are linked In reality, we create unique IDs for entity sets Examples: SSN, ISBN, CS 564 [Spring 2018] - Paris Koutris 42

43 E/R TO RELATIONAL MODEL CS 564 [Spring 2018] - Paris Koutris 43

44 ER MODEL VS RELATIONAL MODEL ER model many concepts: entities, relations, attributes, etc. well-suited for capturing the app requirements not well-suited for computer implementation Relational model has just a single concept: relation world is represented with a collection of tables well-suited for efficient manipulations on computers CS 564 [Spring 2018] - Paris Koutris 44

45 TRANSLATION Basic cases: entity set E -- > relation with attributes of E relationship R -- > relation with attributes being keys of related entity sets + attributes of R Special cases: combining two relations weak entity sets is-a relationships CS 564 [Spring 2018] - Paris Koutris 45

46 ENTITY SET TO RELATION Person name address ssn Person(ssn, name, address) CREATE TABLE Person (ssn CHAR(11) PRIMARY KEY, name CHAR(40), address CHAR(50)) CS 564 [Spring 2018] - Paris Koutris 46

47 ENTITY SET TO RELATION price name category Product Product(name, category, price) CREATE TABLE Product (name CHAR(40), category CHAR(20), price REAL, PRIMARY KEY(name, category)) CS 564 [Spring 2018] - Paris Koutris 47

48 RELATIONSHIP TO RELATION price name category ssn name Product buys Person date address Product(name, category, price) Person(ssn, name, address) Buys(prodname, prodcategory, ssn, date) CS 564 [Spring 2018] - Paris Koutris 48

49 MANY-ONE RELATIONSHIPS price name category name stockprice Product makes Company No need for a Makes relation; instead modify Product: Product(name, category, price, company_name) Company(name, stockprice) CS 564 [Spring 2018] - Paris Koutris 49

50 MANY-MANY RELATIONS price name category ssn name Product buys Person date address Product(name, category, price, ssn) What is wrong here? CS 564 [Spring 2018] - Paris Koutris 50

51 RELATIONSHIP TO RELATION: SQL CREATE TABLE Buys (prodname CHAR(40), prodcategory CHAR(20), ssn CHAR(11), date DATE, PRIMARY KEY(prodname,prodcategory,ssn) FOREIGN KEY (ssn) REFERENCES Person, FOREIGN KEY (prodname, prodcategory) REFERENCES Product(name, category)) CS 564 [Spring 2018] - Paris Koutris 51

52 WEAK ENTITY SETS Team affiliation University number sport name Team(number, affiliated-university, sport) Affiliation does not need a separate relation! Attribute name needed as part of the key CS 564 [Spring 2018] - Paris Koutris 52

53 WEAK ENTITY SETS The relation for a weak entity set must include: attributes for its complete key (including those in other entity sets) its own, non-key attributes A supporting (double-diamond) relationship is redundant and produces no relation CS 564 [Spring 2018] - Paris Koutris 53

54 SUBCLASSES: OPTION 1 price name category Product version ISA age group Software Product Educational Product Product(name, category, price) SoftwareProduct(name, category, price, version) EducationalProduct(name, category, price, age-group) CS 564 [Spring 2018] - Paris Koutris 54

55 SUBCLASSES: OPTION 2 price name category Product version ISA age group Software Product Educational Product Product(name, category, price) SoftwareProduct(name, version) EducationalProduct(name, age-group) CS 564 [Spring 2018] - Paris Koutris 55

56 SUBCLASSES: OPTION 3 price name category Product version ISA age group Software Product Educational Product Product(name, category, price, version, age-group) Use NULL to denote that the attribute makes no sense for a specific tuple CS 564 [Spring 2018] - Paris Koutris 56

57 SUBCLASSES RECAP Three approaches: 1. create a relation for each class with all its attributes 2. create one relation for each subclass with only the key attribute(s) and attributes attached to it 3. create one relation; entities have null in attributes that do not belong to them CS 564 [Spring 2018] - Paris Koutris 57

Conceptual Design with ER Model

Conceptual Design with ER Model Lecture #2 1/24/2012 Jeff Ballard CS564, Spring 2014, Database Management Systems 1 See the Moodle page Due February 7 Groups of 2-3 people Pick a team name Homework 1 is