Normalization in DBMS is called the process of organizing data in order to stop any redundancy from occurring.
DBMS stands for database management system. It is a software package within computer programs. It is meant to develop the databases for various applications. It can work to make things easier for database administrators. One can control the creation, maintenance, and use the database. Data access, data integrity, concurrency control, recovering databases after failures, and restoring backup files are all part of the database management system.
During the design phase of the database management system there can be redundancy issues. Under relational DBMS one tries to minimize the redundancy. This process is called normalization. The goal of normalization is to "decompose" how things are related to each other. It will find the anomalies or issues within the database to ensure that they are less redundant. It could be that additions, modifications, and deletions are made to ensure the normalization process. Edgar F. Codd was the inventor of the relational model and the concept of normalization. Codd first discussed this concept in 1970. At the time computers were for governments and corporations that could afford it. It was not a mainstream concept and thus those who do not study computers or database management systems will not understand that this concept even exists.
There is also the concept of denormalization. This is obviously the opposite of normalization, but it can be used with it. This is because selective denormalization can help with performance.
Codd's model is no longer used as it was in the 70s. It also means that denormalization does not adhere to the Codd model such as the 3NF. The information needed from this is that one can now understand what the DBMS is and why you might use normalization.