NORMALIZATION IN DATABASE MANAGEMENT

SYSTEMS

Course Name: Database Management

Systems
Course Code :
Date CSE311
: 01-12-
2024
GROUP MEMBERS

Raduan Ahmed 221-15-4977

Md.Ariful islam 221-15-
4651
Faisal Islam 221-15-
4178 Shovon 221-15-
Shaikh Sami
Hasibul Hasan5487
221-15-5600
 Introduction to
Normalization :
• What is Normalization?
⚬ In a Database Management System (DBMS), "normalization" refers to the
process of organizing data within a database by structuring tables and
establishing relationships between them, the process of organizing data to
reduce redundancy and improve data integrity.

Importance: Normalization has a lot of importance in databases. Some of this :

⚬ Reduces duplication of data.
⚬ Ensures consistency and accuracy.
⚬ Enhances database efficiency.
OBJECTIVES OF NORMALIZATION :

Minimize Data Redundancy:

• Purpose: Reduce repeated data in multiple places.
• Benefit: Saves storage space and makes data management more efficient.

Eliminate Insertion, Update, and Deletion Anomalies:

• Insertion : Problems occurring when new data can't be added without adding redundant
data.
• Update : Errors arising when multiple copies of the same data need to be updated
consistently.
• Deletion
Ensure Data:Dependencies
Issues when deletion
MakeofSense:
data inadvertently removes useful information.
• Purpose: Organize data such that relationships between different pieces of data are logical
and meaningful.
• Benefit: Enhances data integrity and avoids unpredictable results in queries.
NORMAL FORMS OVERVIEW
1NF: BCNF:
Remove repeating groups. Stricter version of 3NF.

2NF: 4NF:
Remove partial Address multi-valued
dependencies.
dependencies.

3NF: 5NF:
Remove transitive Address join
dependencies. dependencies.
FIRST NORMAL FORM (1NF) Product_Table

Product_ID Color Price

This table is not in first normal(1NF) 1 Green 210$

from because the "Colour" column
contains multiple values 2 Black,Red 155$

3 Blue 69$

Product_Table
ABOUT
Product_ID Color Price

1 Green 210$ A table is in 1NF if:

• Each column contains atomic
2 Black 155$ values.
• There are no repeating groups.
2 Red 155$

3 Blue 69$
SECOND NORMAL FORM (2NF) Table_Purchare_Details

Customer_ID Store_ID Location

This table has a composite primary key
(i.e. Customer_ID, Store_ID). The non key 1 1 Patna
attribute is location. In this case location
depends on Store_ID which is part of the 1 3 Nodia
primary Key
2 1 Patna

Product_Purchase Product_Store ABOUT

Customer_ID Store_ID Store_ID Location A table is in 2NF if:
• It is in 1NF.
1 1 1 Patna
• All non-key attributes are fully
functionally dependent on the
1 3 3 Nodia
primary key.

2 1
THIRD NORMAL FORM (3NF) Step : Identify the Transitive
Dependency
• Dept_Manager depends on
Dept_ID.
• Dept_ID depends on the
• primary key Emp_ID.
 BOYCE-CODD NORMAL FORM (BCNF)

BCNF is a stricter version of

Third Normal Form (3NF).

Key Concepts:
• Functional Dependency (FD): A relationship where one attribute or a
set of attributes uniquely determines another attribute.
• Superkey: A set of attributes that uniquely identifies a record in a table.
• Candidate Key: A minimal superkey, meaning no attribute can be
removed without losing the uniqueness.
Functional Dependencies:
• Student_ID, Course_ID → Instructor (The combination of
Student_ID and Course_ID determines the Instructor).
• Course_ID → Instructor (A course is taught by only one
instructor).
Step 1: Ensure the Table is in 3NF
Step 2: Identify Functional
Dependencies
Step 3: Identify Candidate Keys and
Superkeys
Step 4: Check Functional
Dependencies for Superkeys
 Advantages of Normalization
Normalization has many advantages, including:

• Improved database organization: Normalization organizes a database's tables and columns, and ensures the
database's integrity.

• Reduced redundancy: Normalization can eliminate duplicate data and reduce null values.

• Improved data analysis: Normalization can help avoid data quality issues and improve data analysis.

• Enhanced data security: Normalization can increase data security.

• Easier to use and accurate data: Normalization can make data easier to use and accurate.

• Consistent tables: Normalization can make tables consistent, making them easier to access and understand.

• Faster finding, sorting, and indexing: Because tables are more compact, finding, sorting, and indexing can be quicker.

• Helps models converge more quickly: In machine learning, normalization can help models converge more quickly
during training.

• Helps models infer better predictions: In machine learning, normalization can help models infer better predictions.
 DisAdvantages of Normalization
Data normalization has several disadvantages, including:

• Query performance: Normalization can make querying data more complex and slower, especially in large databases. This is
because normalization requires complex joins between tables to retrieve data.

• Storage space: Eliminating redundancy can increase storage requirements because data is spread across multiple tables.

• Indexing: Indexing doesn't work as efficiently because joins are required. This also makes read times slower.

• Data integrity: Maintaining information integrity can be a challenge.

• Design complexity: The design process can be complex.

• Limited feature variability: Normalization may not provide major benefits if the dataset features are already similar in scale or
have limited variability. Normalizing such datasets can lead to data distortion and negatively impact model performance.

• Tables contain codes: Rewritten data is saved as lines of numbers instead of actual data, so tables contain codes rather than
actual information.
 When to Use Denormalization:

• Performance Optimization: When

DENORMALIZATION queries are slow due to multiple
joins.

Denormalization is the • Reporting Systems: Often used in

process of combining data warehouses for faster data

normalized tables into retrieval.

fewer tables to improve • OLAP systems: Analytical

databases benefit from
query performance.
denormalization, which speeds up
read-heavy Improved
Advantages: operations.Query
Performance and simpler Queries.
Disadvantages: Data redundancy
and Maintenance Complexity.
REAL-WORLD APPLICATIONS

E-commerce Databases:
• Normalization helps store customer, product, and order data separately, avoiding redundancy and
ensuring data integrity.
• Denormalization speeds up product searches and order queries. Products might be stored with
prices and stock information in one table for fast access.

Banking Systems:
• Normalization ensures efficient storage and management of accounts, transactions, customer
Your paragraph text
data, and branch details without redundancy.
• Denormalization is used for reporting purposes, where balances and recent transactions may be
stored together for faster reporting.
THANK YOU