Introduction to Normalization & Dependencies

CS403 Database Management Systems Lecture 19 delivers a rigorous examination of relational database normalization and functional dependency theory, forming the conceptual bedrock of logical database design at the Virtual University of Pakistan. Grounded in authoritative literature including Catherine Ricardo's Database Systems: Principles, Design and Implementation and Raghu Ramakrishnan and Johannes Gehrke's Database Management Systems, this module explains why unnormalized relational tables degrade system integrity and performance. Without systematic normalization, databases suffer from severe operational flaws: data redundancy, wasted secondary storage, and debilitating update anomalies, including insertion anomalies (inability to record facts without unrelated context), deletion anomalies (unintended loss of critical data when deleting a record), and modification anomalies (data inconsistencies when updating duplicated attributes). The curriculum establishes functional dependency as the core mathematical relationship between attributes: given relation R, attribute set B is functionally dependent on attribute set A (denoted A determines B) if each distinct value of A uniquely identifies exactly one value of B across all tuples. Attribute set A functions as the determinant, while B represents the dependent. Understanding functional dependencies enables database architects to systematically identify table keys: any determinant that uniquely determines all attributes in a relation constitutes a super key, while a minimal super key containing no extraneous attributes qualifies as a candidate key. The lecture rigorously presents Armstrong's Axioms, a sound and complete set of inference rules used to derive the complete closure of functional dependencies: Reflexivity (if B is a subset of A, then A determines B, representing trivial dependencies), Augmentation (if A determines B, then AC determines BC), and Transitivity (if A determines B and B determines C, then A determines C). Building upon these primary axioms, secondary inference rules are examined: Decomposition or Projectivity, Additivity or Union, and Pseudotransitivity. Finally, the lecture introduces First Normal Form (1NF), defining a relation in 1NF if and only if every attribute domain contains only atomic, indivisible values, completely eliminating repeating groups, multivalued fields, and composite structures. For students preparing for CS403 midterm examinations, solved past papers by Moaaz and Waqar Siddhu, and chapter-wise solved practice MCQs, mastering functional dependencies and 1NF conversion is essential for academic excellence and enterprise database design.