National University of Computer and Emerging Sciences
Lahore Campus
Database Systems (CS2005) Final Exam
Date: Thu, 23 May 2024 Total Time (Hrs.): 3
Course Instructor(s) Total Marks: 70
ZA, MN, SF, AA, HI, MM, IR Total Questions: 9
___________ __________ _____________________
Roll No Section Student Signature
Do not write below this line.
Note: Please ensure that you attempt all questions and their respective parts in the given order.
SOLUTION
CLO # 2: Design conceptual, logical and physical database schemas using different data models.
Q. No 1: Map the following ER/EER Diagram into a relational model and specify all the constraints
including primary key, foreign key, not null, and unique. [10]
CLO # 2: Design conceptual, logical and physical database schemas using different data models.
Q. No 2: Draw an ER/EER diagram (using notation discussed in lectures) for the following
requirements of a simple database of National Hockey League (NHL). Specify all constraints that
should hold on to the database and state any assumptions you make. [10]
The NHL has many teams, each team has a name, a city, a coach, a captain, and a set of players, each
player belongs to only one team, each player has a name (first, last, middle), a position (such as left
wing or goalie), and a skill level. There are two types of players, beginners, and experts. For
beginners the number of matches per day for training is more than the experts one. The trainers are
just required for beginners and 90% attendance is required in training sessions to remain in the
Spring 2024 FAST School of Computing Page 1 of 7
National University of Computer and Emerging Sciences
Lahore Campus
team. Team captain is also a player, a game is played between two teams (referred to as host_team
and guest_team) and has a date (such as May 11th, 1999) and a score (such as 4 to 2). A player can
also get an injury during a game. A description or record of injury is also maintained for each player.
CLO # 3: Identify FDs and resolve database anomalies by normalizing database tables.
Q. No 3: Consider the following schedule of four transactions T1, T2, T3, and T4.
S: r1(X), r4 (Z), w1(Z), r3 (X), r3 (Y), w1 (X), w3 (Y), r2 (Y), w2 (Z), w2 (Y).
Draw the serializability (precedence) graph for this schedule. State whether this schedule is (conflict)
serializable or not. If the schedule is serializable, write down the equivalent serial schedule(s)
otherwise explain why it is not. [5] Z
T1 T2
Ans: It is conflict-serializable and
X
equivalent serial schedules are Z Z Y
T4T3T1T2 and T3T4T1T2.
It is also view-serializable. T4 T3
CLO # 3: Identify FDs and resolve database anomalies by normalizing database tables.
Q. No 4: Consider a relation R (A, B, C, D, E), with the set of FDs F= {C → D, B → ACE, AD → B}. Find
all possible keys (i.e. candidate keys) of this relation? Prove it. [5]
Ans: Keys are Ans: B, AC, and AD.
CLO # 3: Identify FDs and resolve database anomalies by normalizing database tables.
Q. No 5: Consider the relation schema R (A, B, C, D, E), with FDs F= {ABC → D, AD→ B, E → C, A→ B,
D→ A, BC → A}. Find a minimal cover of F (i.e. Fc). [5]
Ans: Fc= {ABC → D, AD→ B, E → C, A→ B, D→ A, BC → A}
i.e. Fc= {BC → D, E → C, A→ B, D→ A}.
CLO # 3: Identify FDs and resolve database anomalies by normalizing database tables.
Q. No 6: Consider the relation R (A, B. C, D, E), with FDs {A→ B, BC → D, E → C, D→ A}. State which
of the following decompositions of R relation are lossless decomposition. Prove it. [5]
a. R1(C, E), R2(A, B), and R3(A, C, D).
b. R1(C, E), R2(A, B), and R3(A, D, E).
Ans: Keys of R are AE, BE, and DE.
a. Lossy decomposition. R1(C, E), R2(A, B), and R3(A, C, D); Only one cond. is true i.e. R2∩R3→R2.
b. Lossless decomposition. R1(C, E), R2(A, B), and R3(A, D, E); R1∩R3→R1 & R2∩R3→R2.
CLO # 3: Identify FDs and resolve database anomalies by normalizing database tables.
Q. No 7: Consider the relation schema R (A, B, C, D, E, G), with FDs F= {AB → C, CD → EG, EG → A, BC
→ D}. Keys of this relation are AB, BC, and BEG. Identify the best normal form that R satisfies (1NF,
2NF, 3NF, or BCNF). Justify your answer. If R is not in BCNF, decompose it into a set of BCNF relations
and show your steps. Indicate which dependencies if any are not preserved by the BCNF
decomposition. [5]
Ans: HNF=3NF as FD2/FD3 violates BCNF.
BCNF Schema is R1(B, C, D), R2(C, D, E, G), R3(E, G, A). FD1: AB→C is lost.
Spring 2024 FAST School of Computing Page 2 of 7
National University of Computer and Emerging Sciences
Lahore Campus
Consider the following part of Library database for the next questions:
The NUCES library keeps records of the different reports submitted by students to the department,
such as FYP Reports, MS thesis, and PhD thesis. Recently, the library added a new module, Report
Rating, that allows library users to rate and review the reports in the library database. This feature is
beneficial for students looking for quality reports on a subject.
The Rating module includes Users, Reports, and Ratings. Users are identified by their unique ID, and
have username, and gender. Each Report has a title, unique identifier (ReportID), year (publication
year), and Subject (which can be a Database, Operating system, Algorithms, etc).
The authors of a Report are stored in the Report_Author table, as a report (such as an FYP report)
can have multiple authors. AuthorID is a foreign key and refers to UserID in the User table. The
Rating Table links Users and Reports and stores the user ratings and reviews on different Reports.
The user gives a rating on a scale of 1-5.
USER
UserI UserNa Gend REPOR
D me er T
1 Ali male ReportI
Title Subject Year
2 Ahmed male D
3 Hamza male Distribute
Databas
femal 1 d DB for 2010
4 kiran e
e big data
femal Machin
5 Maria Fraud
e e
2 detection 2023
Learnin
using AI
g
Advanced Data
3 leftist Structur 2001
trees e
Applicatio Machin
ns of e
4 1994
Neural Learnin
Networks g
Data
5 AVL Trees Structur 2000
e
REPORT_AUTH RATIN
OR G
ReportID AuthorID ReportI UserI UserRati UserReview
1 1 D D ng
1 2 1 1 4 Excellent
2 3 Highly
3 4 1 2 5 recommend
4 5 ed
5 1 3 2 3 Good read
4 4 4 Classic
Spring 2024 FAST School of Computing Page 3 of 7
National University of Computer and Emerging Sciences
Lahore Campus
literature
Not my cup
5 5 2
of tea
Spring 2024 FAST School of Computing Page 4 of 7
National University of Computer and Emerging Sciences
Lahore Campus
CLO # 4: Use SQL for database definition and manipulation in any DBMS.
Q. No 8: Consider the database state given above and for each of the following queries, give the
output for the database state given above. [10]
a. SELECT u.username FROM User AS u
WHERE u.UserID IN (SELECT a.AuthorID FROM Report_Author AS a
WHERE a.AuthorID IN (SELECT b.UserID FROM Rating AS b
WHERE a.ReportID = b.ReportID));
Answer:
Username
Ali
Ahmed
b. SELECT b.ReportID, b.Title , COUNT(a.AuthorID) AS C
FROM Report b LEFT OUTER JOIN Report_Author a ON b.ReportID = a.ReportID
GROUP BY b.ReportID, b.Title;
Answer:
1 Distributed DB for big data 2
2 Fraud detection using AI 1
3 Advanced leftist trees 1
Applications of Neural 1
4
Networks
5 AVL Trees 1
c. π UserID, UserName (σ UserRating<3 (RATING ⨝Rating.userID=user.userID USER))
userID UserName
5 Maria
d. Subject ℑ COUNT(Subject) (RATING * REPORT)
Subject C
Database 2
Data Structure 2
Machiene Learning 1
Spring 2024 FAST School of Computing Page 5 of 7
National University of Computer and Emerging Sciences
Lahore Campus
CLO # 4: Use SQL for database definition and manipulation in any DBMS.
Q. No 9: Consider the above database for the following problems. [15]
a. Write SQL and RA statements to list the names and IDs of the Authors who have only written a
Report on the Subject “Database”.
SELECT A.AuthorID, U.name
FROM ReportAuthor A JOIN Report R ON A.ReportID = R.ReportID JOIN User U ON A.AuthorID
= U.userID
WHERE R.Subject = 'Database' AND A.AuthorID NOT IN (
SELECT A1.AuthorID
FROM ReportAuthor A1 JOIN Report R1 ON A1.ReportID = R1.ReportID
WHERE R1.Subject <> 'Database' );
DBAuthors = π{AuthorID} (ReportAuthor * σSubject='Database'(Report))
NonDBAuthors = π{AuthorID} (ReportAuthor *{σ{Subject<>'Database'} (Report)})
AuthorNames = π{AuthorID, name} (User ⨝ OnlyDBAuthors)
OnlyDBAuthors = DBAuthors – NonDBAuthors
b. Write SQL and RA statements to list the names of the male users who have reviewed all the
Reports rated 3 or above by the user “Damam Shah” with UserID=101.
SELECT UserNAME
FROM USER U
WHERE Gender = ‘Male’ and
NOT EXISTS (
(SELECT ReportID FROM Rating r
WHERE rating>=3 and r.UserID=101)
EXCEPT
(SELECT ReportID
FROM Rating R
WHERE U.UserID= R.USerID) )
U1 π ReportID (σ rating>=3 and r.UserID=101Rating)
U2 π UserID, ReportID (Rating)
U3 (U2 ÷ U1)
Result π UserName {U3 * σ gender=’Male’ User }
Spring 2024 FAST School of Computing Page 6 of 7
National University of Computer and Emerging Sciences
Lahore Campus
c. Create a View that lists the ReportID, Title, MaximumRating, and the number of reviews received
for the Report, but only for the Reports with a minimum rating above 3 and written by 2 or more
authors.
CREATE VIEW HighRatedReports AS
SELECT R.ReportID, R.Title, MAX(Rating.UserRating) AS MaximumRating,
COUNT(Rating.ReportID) AS NumReviews
FROM Report R JOIN ReportAuthor RA ON R.ReportID = RA.ReportID JOIN Rating ON
R.ReportID = Rating.ReportID
GROUP BY R.ReportID, R.Title
HAVING MIN(Rating.UserRating) > 3 AND COUNT(DISTINCT RA.AuthorID) >= 2;
Spring 2024 FAST School of Computing Page 7 of 7
