DBMS Concepts

Entity Relation Model Armstrong’s Axioms X+ Algorithm of Bernstein

• Entity: Distinguishable from other 1. If 𝑌 ⊂ 𝑋 ⊂ 𝑈 then 𝑋 → 𝑌, reflexive Set N=0 and 𝑋(0) = 𝑋(𝑔𝑖𝑣𝑒𝑛)
objects 2. If 𝑋 → 𝑌 and 𝑍 ⊂ 𝑈 then 𝑋𝑍 → 𝑌𝑍, augmented 1. ∀𝐴 → 𝐵 in f such that 𝐴 ⊂ 𝑋 and 𝐵 ∉ 𝑋.then
• Entity Set: Group of Entities 3. If 𝑋 → 𝑌 and 𝑌 → 𝑍 then 𝑋 → 𝑍, transitive 𝑋 = 𝑋∪𝐵
• Record: Each Row of Dataset 2. Repeat step two until no 𝐴 → 𝐵 such exists
• Attributes: Column-ID
Inference Rules Elementary and Redundant FD
• One - One: 1. {𝑋 → 𝑌, 𝑋 → 𝑍} then 𝑋 → 𝑌𝑍, Union Rule 𝑋 → 𝑌 is Elementary if ∄(𝑋 ′ ⊂ 𝑋 𝑎𝑛𝑑 𝑋 ′ → 𝑌)
2. {𝑋 → 𝑌, 𝑊𝑌 → 𝑍} then 𝑋𝑊 → 𝑍 F is a Redundant FD if (𝐹 − 𝑓)+ = 𝐹 +
3. {𝑋 → 𝑌, 𝑍 ⊂ 𝑌} then 𝑋 → 𝑍 In such cases partial dependency exists
• Many - One: 4. {𝑋 → 𝑌𝑍} then 𝑋 → 𝑌, 𝑋 → 𝑍
Lossless Join Property
• One - Many: Keys If 𝑟 = 𝜋𝑅1 (𝑟) ⋈ 𝜋𝑅2 (𝑟)
• Super: A combination of fields by which a row is Lossless if 𝑅1 ∩ 𝑅2 → 𝑅1 − 𝑅2 𝑜𝑟 𝑅2 − 𝑅1 𝑒𝑥𝑖𝑠𝑡𝑠 𝑖𝑛 𝑓
uniquely identified. If u add any column to primary key
• Many -Many: it becomes a super key. 1st Normal Form(1NF)
A
• Primary key: Any key which can uniquely identify As per the rule of first normal form, an attribute (column)
rows of the current table of a table cannot hold multiple values. It should hold only
• Subset Rel: Is A
B is a
• Candidate key: The minimal super key. Individual
subset atomic values.
of A
B columns in a table which qualifies for uniqueness of all
rows Boyce Codd Normal Form(BCNF)
Models • Foreign Key: Field or collection of fields which A relation is in BCNF 𝑖𝑓 𝑋 → 𝑌 𝑒𝑥𝑖𝑠𝑡𝑠 & 𝑌 ∉ 𝑋
• Network: ER model restricted to uniquely identifies a row of another table. then X is a key(candidate or super)
Binary, Many-One Relationships,
To break a many many relation 3NF or Third Normal Form Prime Attribute
𝑋 → 𝐴 is in 3NF if 𝐴 ∉ 𝑋 and either X is a key of A attribute which is a member or a subset of the key
insert temp between them and
insert 2 many-one relations relation or 𝐴 is a prime attribute.
• Hierarchical: Kind of Decision Transformation removes Partial And Transitive 2nd Normal Form(2NF)
Tree Dependency • Table is in 1NF (First normal form)
• No non-prime attribute is dependent on the proper subset of
Basic Relational Algebra Operators any candidate key of table.
• Union: 𝑅 ∪ 𝑆 Set of Tuples either in R or S both. R,S must have same number of attributes Partial Dependency is Removed
• Set Difference: 𝑅 − 𝑆 Set of Tuples present in R but not in S. Same attributes
• Cartesian Product: 𝑅 × 𝑆, Let R,S have arity(no of attributes) k1,k2. Then RxS is set of tuples Additional Relational Algebra Operators
st
with arity k1+k2 with 1 k1 components form tuple of R and last k2 components tuples of S • Intersection: 𝑅 ∩ 𝑆 = 𝑅 − (𝑅 − 𝑆) Set of Tuples both in R
• Projection: 𝜋𝑖1,𝑖2,…,𝑖𝑁 (𝑅) Set of tuples with attributes 𝑖1 , 𝑖2 , … , 𝑖𝑁 with same content with that of or S both. R,S must have same number of attributes
R • Quotient: 𝑅 ÷ 𝑆 let arity of R is r and S is s and 𝑟 > 𝑠, set
• Selection: 𝜎𝐹 (𝑅) set of tuples satisfying criteria F {Select C1,C2 from T1 where C3>2} of tuples ‘t’ with arity 𝑟 − 𝑠 st ∀ tuples ‘U’ in S, the tuple
𝑡𝑢(concat) is in R
Core Diagram Terminologies • 𝜽 Join: 𝑅 ⋈ 𝑖𝜃𝑗 𝑆, Set of Tuples that is in the subset of
• Cardinality: No of Tuples 𝑅 × 𝑆 satisfying 𝑖𝜃𝑗
• Natural Join: 𝑅 ⋈ 𝑆, Subset of 𝑅 × 𝑆 in which only those
• Arity: No of Attributes
tuples with same values for same attributes in R,S are
• Tuples: Represents Rows represented in same column

Type of Joins
• Natural Join: (or Inner Join) when tuples with
same values for same attributes in R,S are
represented in the same column
• Left Join: Takes values of all tuples of left
join, wherever same values for same attributes
in R,S are represented in the same column, if
there doesn’t exists a value in R, it will print
NULL
• Right Join: here R and S is interchanged from
above