Data Ware House
Normalization it’s a process of splitting data into 2 or more entities to reduce data redundancy.
1. A database table must contain no repeating groups
2. All non-key attributes of a table must rely on the entire key of the table. 3. All non-key fields must depend solely on the table’s primary key.
First Business Normal Form: removes repeating groups to another entity. This entity takes its name and primary (compound) key attributes, from the original entity and forms the repeating group. (Repeating groups of data are moved to another table). Second Business Normal Form: removes attributes that are partially dependent on the primary key to another entity. The primary (compound) key of this entity is the primary key of the entity in which it originally resided, together with all additional keys on which the attribute is wholly dependent. (Every non-key attribute is dependent on key attribute) Eliminate redundant attributes, which can be of any of the following: 1. Different attributes that represent the same concept
2. Attributes in different entities that share a common theme. 3. Attributes which do not have a value for every instance of an entity. Number of candidates in a primary key varies with RDBMS for Oracle 8I –32. Sum of sizes of candidate keys should not exceed 1/3 of block size. Third Business Normal Form: removes attributes that are not dependent at all on the primary key to another entity where they are wholly dependent on the primary key of that entity. (There should not be any transitive dependency among the non-key attributes; i.e. non-key attributes must be mutually independent). Eliminate any attributes that are dependent upon values of other attributes in addition to the primary key (some times known as Transitive Dependency). Fourth Business Normal Form: removes attributes that are dependent on the values of the primary key or that are optional to a secondary entity where they wholly depend on the...
Please join StudyMode to read the full document