Mar 8, 2020

Note Class 12 : Chapter 2 DATABASE

 \





Basic terms used in database
1)   Data: -A raw fact about anything which does not give any complete meaning.
2)   Information: - The processed data which gives some or complete meaning is called information.
3)   Database: - An organized collection of data which are shared and used for multiple purposes is called database. Example: - data of SLC result, Telephone diary etc.
4)   Database Management System (DBMS): -The software collection which helps to manage the database is called DBMS. E.g.:- FoxPro, Oracle, Ms-Access etc.
5)   Field: -A field is a piece of information about an element; element may be a person, student or employ etc. in figure SN, NAME and PHONE NO are fields.
SN
NAME
PHONE NO
1
ALK
9844128670
2
RAM JANAKI SOFT
9813580600
6)   Record: - A collection of related field is called record. A single row containing of one data is a record.

1
ALK
9844128670

7)   Domain: -A domain is a set of values from which the actual values appearing in a given relation, for attribute customer name, the domain is the set of all customer names.
Customer
Customer_name
Account no
Balance
Raju
A-001
4,44,623
Sita
A-002
10,23,245
Shankar
A-007
35,25,006


8)   Tuple: -The row of relation (table) is called tuple. In this relation below, there are 3 tuple.

Customer
Customer_name
Account no
Balance
Raju
A-001
4,44,623
Sita
A-002
10,23,245
Shankar
A-007
35,25,006

2.1.2 Objectives or Importance  of DBMS: - 
The following are the objectives of DBMS.
a)    To provide huge storage or space for relevant data.
b)   To allow easy access to the data for the user.
c)    To provide quick response to user request for any information or data.
d)   To allow updating with the latest modification in the database.
e)    To remove duplicate data.
f)     To allow many users to the database at one time.
g)    To allow the growth of database system.
h)   To provide maximum protection to data from any physical damage and unauthorized access.
2.1.3 Database Model: -
A database model is the method of organizing data and represents the logical relationship among data elements in the database. The most popular database models are:
1. Relational Database Model (RDBMS): -
The relational model was formally introduced by Dr. E.F. Codd in 1990. The relational model represent data in the form of two-dimensional tables called relation, which are made of columns and rows. Each column represents a field, also called an attribute, each row represent a record, also called a Tuple. The domain is a pool of data values from which data is entered into table for database management. A world of real world data is called domain.
Example of Relational data Model: -
The database has 3 tables (relations):
Employee, Department & Project.
EMPLOYEE (E No, E Name, E Add, E Sal, Gen, DNO)
DEPARTMENT (D No, D Name, D Location)
PROJECT ( P No, P Name ,P Location, E No.)
The entity name is shown in UPPERCASE;
The primary key field is underlined;
The attributes are shown in brackets, separated by commas.
EMPLOYEE
E No
E Name
E Add
E Sal
Gen
DNO
E1
AAKASH DANGOL
Kathmandu
1,00,000
Male
2
E2
AAKASH SHRESTHA
Kathmandu
1,00,000
Male
3
E3
ALINA BASTAKOTI
Kathmandu
1,00,000
Female
1
E4
ANJU SHRESTHA
Kathmandu
1,00,000
Female
5
E5
ANU SHRESTHA
Kathmandu
1,00,000
Female
4

DEPARTMENT
D No
D Name
D Location
1
Library
101
2
Laboratory
102
3
Account
103
4
Examination
104
5
Discipline
105

PROJECT
P No
P Name
P Location
E No
1
English book
101
E1
2
Com. Networking
102
E2
3
Billing System
103
E3
4
Grading
104
E4
5
Speaking
105
E5
                                                 
Note: - E No is the primary key for table Employee and the foreign key for the table project similarly; D No. is the primary key for the table of Department and foreign key for the table of Employee. The primary key and the foreign key are the fields that relate tables to each other.


Advantages of Relational Database Model (RDBMS):-
a)    The rules are common in each table and easy to link one table to another.
b)   Normalization of database is possible.
c)    Quick database processing is possible.
d)   It has very less redundancy (unnecessary data).
e)    It enables a computer system to accommodate a variety of file inquires in an efficient manner.
f)     It also helps to add indexes for table.

Disadvantages of Relational Database Model(RDBMS):-
a)    It is complex than other models.
b)   It is confusing as many rules being applied and becomes non-user friendly.
c)    The index portion of the file must be created and maintained along with the file records.
d)   In some cases the index portion of the file may be larger than the file with the file records.
e)    The file index must be searched sequentially before the actual file records are obtained, resulting in wastage of time.

2. Network Model:-
In network model, data are represented by a collection of records and relationships among data are represented by links. A link is an association between records. The records are recognized as arbitrary graph. Network model is still popular on powerful mainframes.
Advantages of Network Model:-
a)    This model is more flexible.
b)   It reduces redundancy.
c)    Searching is faster.
Disadvantages of Network Model:-
a)    It is very complex type of database model.
b)   It needs long programs to handle the relationship.
c)    Pointers needed in the database model increases overhead of storage.
Less security in comparison to hierarchical model because it is open to all.

figure: Network Model


3.  Hierarchical Model: -
The Hierarchical data model organizes data in a tree structure. There is a hierarchy of parent and child data segments. Data structure implies that a record can have repeating information, generally in the child data segments. Data in a series of records, which has a set of field values attached to it. It collects all the instances of a specific record together as a record type. These record types are the equivalent of rows. To create links between these record types, the hierarchical model uses parent child relationships. In a hierarchical database the parent child relationship is one to many. This restricts a child segment to having only one parent segment.
Advantages of Hierarchical Model: -
a)    It is the easiest model of database.
b)   A database owner is more secured.
c)    Searching is fast and easy, if parent are known.
d)   Very efficient in handling ‘one to many’ relationship.
Disadvantages of Hierarchical Model: -
a)    It is old fashioned, outdated database model.
b)   Cannot handle ‘many to many’ relationship.
c)    It is non-flexible database.
d)  Increases redundancy. 

4.  E-R (Entity Relationship) diagram
The Entity Relationship (ER) Data Model is a detailed logical representation of the entities, associations and data elements for a system. A graphical representation of the ER data model is known as ER Diagram.
An entity is a person, place, object, event or concept in the user environment about which the organization wishes to maintain data. An entity has its own identity and attributes that distinguish it from other entities. Different entities have different relationships with one another.
The basic symbols used in ER diagram are:

2.1.4 Concept of Normalization:-

The process of breaking the complex relation into simple relation is called normalization. It reduces redundancy using the principle of non-lose decomposition. Non-loss decomposition is the reduction of a table to smaller tables without loss of information. Normalization is needed because it represents a database in normal form to avoid undesirable things.

·       Objectives or importance  of Normalization are 
        a)    It reduces redundancy.
        b)   It improves faster storing and indexing.
        c)    It simplifies the structure of database.
        d)   Dependence between the data is identified.
        e)    Protects from unauthorized users and secures the data in database.
        f)     Removes anomalies for database activities.
        g)    The database model is made more flexible and easier to maintain.


     

2.1.5 Types of normalization 1 NF, 2 NF, 3 NF
1. First Normal form (1 NF): -A form or table to become 1 NF it’s all attributes must be atomic. That is, there can be exist no repeating groups in an attribute. In every tuple of the relation each attribute must have a value.
Member
Mem_code
Mem_Name
Class
Book



Code
Name
Issue-date
Due-date
M001
SUSHMITA SHRESTHA
11
B0012
Computer
2075/05/06
2075/05/26
M002
SUZAL GURUNG
12
B0014
Link Eng
2075/05/06
2075/05/26
M003
THIR PRASAD DAHAL
12
B0015
Math
2075/05/06
2075/05/26
M004
TRILOK LAMA BAMJAN
11
B0034
Physics
2075/05/06
2075/05/26
M005
YUNIK TAMANG
11
B0032
Chemistry
2075/05/06
2075/05/26
Unorganized Relation

Mem_code
Mem_Name
Class
B_Code
B_Name
Issue-date
Due-date
M001
SUSHMITA SHRESTHA
11
B0012
Computer
2075/05/06
2075/05/26
M002
SUZAL GURUNG
12
B0014
Link Eng
2075/05/06
2075/05/26
M003
THIR PRASAD DAHAL
12
B0015
Math
2075/05/06
2075/05/26
M004
TRILOK LAMA BAMJAN
11
B0034
Physics
2075/05/06
2075/05/26
M005
YUNIK TAMANG
11
B0032
Chemistry
2075/05/06
2075/05/26
Figure:-Normalization in 1 NF
Problem in 1 NF: -
a)    Insert:-In the relation, we cannot enter the details of a member unless he/she has taken a book from library, i.e. can’t enter information without book information.

a)    Delete: - if we want to delete the details of book, we shall have to delete the member’s detail also.
b)   Update:-If a member has taken 5 books from library, then the member’s details are repeated 5 times in the relation (table).

2. Second Normal Form (2 NF):- A relation is in 2 NF if it is in 1 NF and each attribute is fully functionally dependent on the primary key.
Problem in 2 NF: -
a)    Insert: - We can insert the details of member even if book information is not available.
b)   Delete: - We can delete book information without deleting the member’s details.
c)    Update: - We can easily insert or delete member information without redundancy.
Member
Mem_codes
Mem_Name
Class
M001
SUSHMITA SHRESTHA
11
M002
SUZAL GURUNG
12
M003
THIR PRASAD DAHAL
12
M004
TRILOK LAMA BAMJAN
11
M005
YUNIK TAMANG
11

Book
Mem_codes
B_Code
B_Name
Issue-date
Due-date
M001
B0012
Computer
2075/05/06
2075/05/26
M002
B0014
Link Eng
2075/05/06
2075/05/26
M003
B0015
Math
2075/05/06
2075/05/26
M004
B0034
Physics
2075/05/06
2075/05/26
M005
B0032
Chemistry
2075/05/06
2075/05/26
Figure:-Normalization in 2 NF
3. Third Normal Form (3 NF):- A relation is in 3 NF if it is 2 NF and each non-key attribute is fully functionally dependent on the entire primary key, and not on any other key. That is, no transitive dependencies exist among the attributes.
The 3 NF overcomes all the problems of 2 NF.
Member
Mem_codes
Mem_Name
Class
M001
SUSHMITA SHRESTHA
11
M002
SUZAL GURUNG
12
M003
THIR PRASAD DAHAL
12
M004
TRILOK LAMA BAMJAN
11
M005
YUNIK TAMANG
11
                          
Book
B_Code
B_Name
B0012
Computer
B0014
Link Eng
B0015
Math
B0034
Physics
B0032
Chemistry

Issue
Mem_codes
B_Code
Issue-date
Due-date
M001
B0012
2075/05/06
2075/05/26
M002
B0014
2075/05/06
2075/05/26
M003
B0015
2075/05/06
2075/05/26
M004
B0034
2075/05/06
2075/05/26
M005
B0032
2075/05/06
2075/05/26
              
Figure:-Normalization in 3 NF
2.1.6 Structured Query Language (SQL):-          
·       It is a standard relational database language.
·       It was developed at IBM’s San Jose Research laboratory.
·       It is a non-procedure language that is the users specify what must be done, but not how it is to be done.
·       There are four basic operations in SQL: SELECT UPDATE, INSERT and DELETE.
·       The SELECT statement enables users to update the data, insert new data and delete existing data respectively.

2.1.7 Centralized Vs Distributed Database
Centralized database: - The host computer is used to run the DBMS, the application that access the database, and the communication facilities that send and receive data from the user’s terminals. The users access the database through either locally connected or dial up terminals. The terminals are generally dumb.
Advantages of the Centralized Database:-
·       Control: - The Database is easily controlled by higher authority.
·       High security: - The Database is centralized so there is no chance of data modifications.
·       Maintenance: - The Database is maintained easily because it is centrally stored.
·       Speed:-Working speed is more than distributed database.
·       Cost:-The database and device required cost is low.
·       Manpower; - Requires less manpower to operate the database.
Distributed Database: - This is a complex types of database system in which a collection of multiple logically interrelated databases are distributed countries wide or out of geographical boundaries. The computers in distributed system communicate with each other through various communication media, such as high speed networks or telephone lines. These computers do not share main memory or disc.



Advantages of Distributed Database:-
·       Transparency
·       Reliability and availability
·       Improved performance
·       Easier expansion
·       Sharing
·       Efficient and flexible
·       Capacity and increment growth.
2.1.8 Data Security
The protection of data is called data security. It means preventing the loss of data, misuse, disclosure or unwanted modification of data.
Various methods can be taken to ensure the security. Here are some common methods:-
a)    Data may be lost due to infection of virus or due to any accident. So to prevent from such problem, use of backup copies of data.
b)   Use of strong password to prevent unauthorized use of computer or unauthorized access to online files.
c)    Physical prevention, restricting of personal, keeping data under lock and key.
d)   Constant checks of security.
e)    Use of latest operating system.
f)     Use of nice antivirus software.
# Data integrity: - Data integrity refers to validity of data contained in database. Database integrity can be reduced in many ways including input typing errors, hardware malfunctions and data transmission errors. To avoid data integrity errors, database programs should use data validation process, which define acceptable ranges for each field in record. If user tries to input data is out of this range, an error message is displayed.
·       Types of data integrity:-
The two integrity rules are called entity integrity and referential integrity.


1. Entity integrity:-Entity integrity is the rule that no column that is part of the primary key may accept null values.
Entity integrity guarantees that each record will indeed have its own identity. In other words, entity integrity prevents the primary key from accepting null values and ensures that one record can be distinguished from other.

2. Referential integrity:- The Referential integrity rule states that if table A contain a foreign key that matches the primary key of table B, then values of this foreign key either must match the value of the primary key for some row in table B or must be null.

#DBA (Database Administration):-
An information specialist who has responsibility for the database is called a Database Administration (DMA). Ideally, DBA is a mature individual with years of computer experience, a wide diversity of technical abilities, and superior managerial skills.
Ø The duties fall into four major areas: - Database planning, implementation, operation, and security.
Typical responsibilities of a DBA are:-
1)    Helps an organization to decide which department will be responsible for the maintenance and update of each data field in a database.
2)    Assures access to database information to each department that needs it.
3)    Secures databases from an authorized use.
4)    Protects databases from physical harm.
5)    Co-ordinates the work of individuals making file making file modifications, policy changes, and improvements to database.



Some of the advantages of database system in comparison to manual system are:-
1)    Redundancies and inconsistencies can be reduced.
2)    Better services to the users.
3)    Flexibility of data system is improved.
4)    Cost of developing and maintaining system is lower.
5)    Data integrity, security etc can be improved.

Ø Data Dictionary:
A data dictionary is a file which contains meta-data that is data about data. It is also called information system catalogue. It keeps all the information about the database system such as location, size of the database, tables, records , fields, user information, privileges, backup system, and recovery system etc. It also defines the data types for each fields etc. A good data dictionary always ensures the consistency in database.


Ø Data Manipulation Language (DML):
A data manipulation language (DML) is language that enables users to access data as organized by appropriate data model, i.e. DML is used to update the database by adding new data or modify or delete the existing data.
Example:
To insert data to the table “record” which contains data like name, address, telephone
insert into record values(“ALK”, “Kathmandu”, 9844128670);


Ø Data Definition Language(DDL):
A database scheme is specified by a set of definitions expressed by special language called a DDL. The logical structure and files within the database may be defined using DDL. Attributes such as record layouts, fields, key validation can be described using a DDL. Example: A table “record” which contains fields like name, address and telephone can be created with DDL.
Create table record (Name               varchar(50),
                                      Address                varchar(50),
                                      telephone             int);


[Unit-2 Homework Questions]
1)   What is normalization? Explain the normalization process with examples.
2)   Define DML and DDL with example.
3)   Explain Relational database model.
4)   Define database and DBMS. Explain the advantages of Database system over Flat-file system.
5)   Who is database administrator? Explain the duties and responsibilities of DBA?
6)   What are the advantages of distributed database system over centralized database?
7)   Explain the advantages of centralized database management system?
8)   What is data integrity? Why it is important in database design? Explain.

Labels:

0 Comments:

Post a Comment

Subscribe to Post Comments [Atom]

<< Home