Normalization
-------------
to avoid data redundancy
to effectively design db

Level of Normalization
----------------------
each level have certain rules
1NF, 2NF, 3NF, BCNF, 4NF, 5NF, 6NF
mostly happens until 3NF


De-Normalized data
------------------
all data in single file/table

in de-normalizaed database, it may cause
* have insertion anomalies
* have deletion anomalies
* have updation anomalies 

1NF
---
1.Every column should have single values (no x,y)
2. Each row should be unique. Either through a single or multiple columns. Not mandatory to have primary key.

2NF
----
1. Must be in 1NF
2. All non key attributes must be fully dependent on candidate key.
if a non-key column is partailly dependent on cadidate ky (subset of columns forming candtidate key) then split them into separate tables.
3.Every table should have primary key and relationship between te tables should be formed using foreign key


Candidate Key
-------------
* set of columns which uniquely identify a record.
* A table can have multiple candidate keys because there can be multiple set of columns which uniquely identify a record/row in a table.

Non-key columns
---------------
columns which are not part of the candidate key or primary key.

Partial Dependency
------------------
* if your candidate key is a combination of 2 columns (or multiple columns) then every non key column (columns which are not part of the candidate key) should be fully dependent on all the columns.

* if there is any non key column whihch depends only on one of the candidate key columns then this results in partial dependency.


3NF
---
1. Must be in 2NF
2. Avoid Transitive dependencies

Transitive dependency
---------------------
A table has 3 columns namely A, B and C
If A is functionally dependeny on B adn B is functionally dependeny on C then we can say A is functionally dependeny on C.



==================================================
From - https://www.youtube.com/watch?v=rBPQ5fg_kiY
==================================================