A data model is a collection of conceptual for describing –

Data

Data relationships,

Data semantics, and

Consistency constraints.

            In other words, a data model describes the structure of a database.

            Data models are either logical data model or physical data model. Logical data models can further be divided into two categories as given below.     

1.     Object-based logical data models (logical data models).

2.     Record-based logical data models (logical data models).

All of these data models are described in following sub-sections.

 

Data models

1.    Object-based logical data models

Object-based logical data models are used in describing data at the logical and view

levels. Some of the well-known object-based logical data models are-

                        The entity-relationship model

                        The object-oriented model

                        The binary model

                        The semantic data model

                        The functional data model

            Among all these models, entity-relationship model and object-oriented model are described in brief here.

1.     Entity-relationship model

The entity-relationship model is based on a perception of a real world. a real world

consists of a collection of basic objects and relationship among these objects. This model is also referred as the E-R model.

            These basic objects are called entities. They are distinguishable from other entities. For example, each customer is an entity. Also, each account is an entity.

            A set of attributes or properties is associated with each entity to describe it. For example, account number and balance describe one particular account in a bank. The set of all entities of the same type is called the entity set. So, customer can be considered as an entity-set and all individual customers are entities.

            An association among several entities is known as relationship. For example, an Account_Holder relationship associates a customer with each account he/she has. A set of all relationships of the same type is called relationship set.

            Along with entities and relationships, the E-R model represents certain constraints. The contents of a database must conform to these constraints. On important constraint is mapping cardinalities. Mapping cardinalities express the number of entities to which another entity can be associated via a relationship set.

            The overall logical structure of a database is represented graphically by E-R diagram.

            The entity-relationship model is widely used in database design. Chapter 8 and 9 explores this model in detail.

 

Data models

2.     Object-oriented model

The object-oriented model is also based on a collection of objects, like as the E-R

model.

            An object:

            An object contains values stored in instance variables within the object. An object also contains bodies of code that operate on these values. These bodies of code are called methods. In other words, object can be considered as-

            Object = values in instance variables + methods

            A class:

            Objects that contain the same types of values and the same methods are grouped together into classed. A class may be viewed as a type-definition for objects.

relational algebra in dbms

            Sending a message:

            The only way in which one object can access data of another object is by invoking (calling) a method of that other object. This is called sending a message to the object.

            Two levels of data abstraction:

            The call interface of the methods, i.e. method names, of an object defines that object’s externally visible part. The internal parts of the object -  the instance variables and method code – are not visible externally. The result is two levels of data abstraction.

            Consider an object account having instance variable acc_no and balance, and  method pay_interest to add interest to balance. Assume that the bank changes its policy to pay interests. It decides to pay 8% interest instead of 5%. This requires to change the code of pay_interest method only, rather than changing the entire application program. The external interface, i.e. method name, to the objects remains unchanged.

            Distinction among objects:

            Each object has its own unique identity in this model. This identity is independent of th values that it contains. Two objects containing the same values are distinct. This distinction is create and maintained in physical level by assigning distinct object identifiers.

2.    Record-based logical data models

Record-based logical data models are also used in describing data at the logical and view

levels. They are used to specify overall logical structure of the database, and to provide a higher level description of the implementation.

In this model, database is structured in fixed-format records of several types. Each

record type defines a fixed number of fields, or attributes. Each field is usually of a fixed length. This simplifies the physical level implementation of the database.

Some of the well-known record-based logical data models are –

            The relational model

            The network model

            The hierarchical model

1.     Relational database model:

The relational model uses a collection of tables to represent both data

and the relationships among those data. Each table contains multiple columns. Each column has a unique name and a fixed width.

            A banking enterprise can have several tables such as customer, account, loan, branch, employee, transaction, account_holder, borrower, etc. The following figure depicts a sample relational database having three tables – Customer, Account and Account_Holder.

Customer:

CID			NAME		SOCIETY	CITY
C01			RIYA		AMUL AAVAL	ANAND
C02			JIYA		SARDAR COLONY	KARAMSAD
C03			PIYA		MARUTISADAN	VVNAGAR
C04			DIYA		SARAL SOCIETY	ANAND
C05			TIYA		BIRLA GRUH	VVNAGAR
	ANO	BALANCE
	A01	5000
	A02	6000
	A03	7000
	A04	8000
	A05	6000
			Account:

CID	ANO
C01	A01
C02	A02
C03	A03
C04	A04
C05	A05

            Account_Holder:

Here, Customer and Account table represents data about Customer and Accounts. Account_Holder table represents relationships among Customers and Accounts. According to these tables, a customer ‘RIYA’ from city ‘ANAND’ has customerID ‘C01’ and she has Account having account numbers ‘A01’. An Account ‘A01’ has BALANCE Rs. 5000.

            These tables also describe that ‘JIYA’ owns two Accounts-‘A02’ and ‘A04’. Also, an account ‘A04’ is jointly acquired by Customer ‘C02’ and ‘C04’, means, ‘JIYA’ and ‘DIYA’.

Date Conversion in SQL

2.     Network model

Data in network model are represented by collections of records. A record is defined as a collection of logically related data items. Relationships among data are represented by links. Links can also be considered as pointers. The records in a database are organized as collections of arbitrary graphs.

C01

RIYA

AMUL ALVAS

ANAND

                                                                                                A01    5000

C02

JIYA

SARDAR COLONY

KARAMSAD

                                                                                                A02    6000

C03

PIYA

MARUTISADAN

VVNAGAR

                                                                                                A03    7000

C04

DIYA

SARAL SOCIETY

ANAND

                                                                                                A04    8000

C05

TIYA

BIRLA GRUH

VVNAGAR

                                                                                                A05    6000

3.     Hierarchical model

The hierarchical model is simply a special case of the network data model. Data are

represented by collections of records as similar to network model. Relationships among data are represented by links or pointers. The records in a database are organized as collections of trees rather than arbitrary graphs.

            Represents a sample hierarchical database using the same information as in figure. Limited records have been considered here.

BANK

C01

RIYA

AMUL AAVAS

ANAND

C02

JIYA

SARDAR COLONY

KARAMSAD

C05

TIYA

BIRLA GRUH

VVNAGAR

A01

5000

A02

6000

A05

6000

1.    Physical data models

Physical data models are used for higher level description of storage structure and

access mechanism.

They describe how data is stored in database. They represent information such as

record structures, record orderings and access paths.

Some of the well-known physical data models are –

            The unifying model

            The frame memory model

4.    Comparison between data models

DATA MODEL	DATA ORGANIZATION	RELATIONSHIPS	IDENTITY	DATA INDEPENDENCE
E-R	Entities	Relationships	Value-based	yes
Object-oriented	Objects	Recursive object	Record-based	Yes
Relational	Tables	Tables	Value-based	Yes
Hierarchical	Files, Records	Pointers of links	Record-based	Yes
Network	Files, Records	Pointers of links	Record-based	yes