Gate 2018

AI Summary

An AI-generated summary of this video lecture.

The video analyzes a GATE 2018 question asking to identify a relational schema that is in 3NF but not in BCNF. The instructor reviews four schemas: Registration with set-valued courses, Registration with email, Registration with marks and grade, and Registration with credit. He evaluates primary keys and functional dependencies for each. He determines that Schema I violates 1NF due to multi-valued attributes. He finds Schema II satisfies 3NF but fails BCNF due to a dependency on a non-superkey. He identifies Schema III as violating 3NF due to transitive dependency. He identifies Schema IV as violating 2NF due to partial dependency. The final conclusion points to Schema II as the correct answer.

Chapters

1. 0:00 – 2:00 00:00-02:00

   The instructor begins by reading the problem statement visible on the slide: "Which one of the relational schemas above is in 3NF but not in BCNF?" He starts with Schema I: Registration (rollno, courses). He points out that the attribute 'courses' is a set-valued attribute. This typically violates First Normal Form (1NF). Although he briefly writes "BCNF" next to it, he dismisses it. He then shifts focus to Schema II: Registration (rollno, courseid, email). He identifies the primary key as the composite key (rollno, courseid). He lists the non-trivial functional dependencies: rollno, courseid -> email and email -> rollno. On the whiteboard, he writes (Rno, Cid) and (Cid, Email). He explains that while this schema satisfies 3NF conditions, the dependency email -> rollno violates BCNF because email is not a superkey. He marks "Schema 2" with a checkmark.

2. 2:00 – 3:37 02:00-03:37

   Continuing the analysis, the instructor examines Schema III: Registration (rollno, courseid, marks, grade). He identifies the primary key as (rollno, courseid) and notes the dependencies rollno, courseid -> marks, grade and marks -> grade. He draws arrows to visualize the transitive dependency where marks determines grade. Since marks is a non-prime attribute determining another non-prime attribute, he labels this schema as "2NF", indicating it fails 3NF. Next, he analyzes Schema IV: Registration (rollno, courseid, credit). He identifies the primary key as (rollno, courseid) and lists dependencies rollno, courseid -> credit and courseid -> credit. He points out that courseid -> credit is a partial dependency because courseid is a proper subset of the primary key. He labels this schema "1NF". Finally, he confirms that Schema 2 is the only one satisfying the condition of being in 3NF but not BCNF.

The lecture demonstrates a methodical approach to database normalization problems by evaluating functional dependencies against normal form definitions. The instructor uses visual aids like arrows and handwritten notes to track keys and dependencies, effectively ruling out schemas that violate lower normal forms (1NF, 2NF) or BCNF constraints. The key takeaway is distinguishing between partial dependencies (violating 2NF), transitive dependencies (violating 3NF), and dependencies on non-superkeys (violating BCNF). The final conclusion reinforces that Schema 2 is the correct answer because it has a transitive dependency on a non-prime attribute (3NF) but also a dependency on a non-superkey (BCNF violation).