Bangladesh Ordnance Factory (BOF) Assistant Engineer (EEE/ME/CSE) Exam: 2021প্রশ্ন ও সমাধান

All three uniquely identify a tuple (row) in a relation, but they differ in scope and how they're chosen:

Key	Definition	Count per table	Nullable?
Super Key	Any attribute set whose values uniquely identify each tuple — may contain redundant attributes	Many possible	No restriction beyond uniqueness
Candidate Key	A minimal super key — removing any attribute destroys uniqueness	A relation can have several	Must be non-null and unique
Primary Key	The one candidate key the designer selects as the table's main identifier	Exactly one per table	Never null (entity integrity)

Example: STUDENT(Roll_No, Phone, Name, Dept)

• {Roll_No} and {Phone} are each candidate keys — both independently identify a student uniquely.
• {Roll_No} is chosen as the primary key.
• {Roll_No, Name} is a super key but not a candidate key, since Name is redundant (not minimal).

graph TD;
SK["Super Keys (many)"]-->CK["Candidate Keys (minimal super keys)"];
CK-->PK["Primary Key (one chosen candidate key)"];

Exam takeaway: Super Key ⊇ Candidate Key ⊇ Primary Key — every primary key is a candidate key, and every candidate key is a super key, but not the reverse.