COMP2521: Data Structures and Algorithms

Quizzes 14%
Assignments 35%
Labs 11%
Final Exam 40%

Prerequisite: COMP1511 or DPST1091 or COMP1917 or COMP1921 (stolen from above link)

Perhaps because of the organisation of COMP1531 which I did at the same time, COMP2521 felt particularly poorly run. The course content was appealing because it was basically logic with extra steps, but I couldn't really shake the fact the course didn't have that feel good factor. It didn't really help that extra recordings and tutorials were often better resources to learn from than lectures, though there was markedly less work than last term (gauging by the reports of one friend complaining about the workload). The tutors were great - but there's really not much to add apart from the fact that unless you enjoy the content, I would not recommend taking the course as it exists right now.

2x 2hr lectures
1x 3hr Tute/Lab

3/5

Yes

Ashesh Mahidadia

Yes

2/5

Suggested texts are available:
Algorithms in C, Parts 1-4: Fundamentals, Data Structures, Sorting, Searching (3rd Edition)
by Robert Sedgewick, published by Addison-Wesley
Algorithms in C, Part 5: Graph Algorithms (3rd Edition)
by Robert Sedgewick, published by Addison Wesley

2020 T3

86 HD

- Lab exercises, worth 8%
- 2x programming assignments, worth 22% combined
- Midterm programming exam, worth 10%
- Final exam with theory and practical sections, worth 60%

COMP1511 is the only formal prerequisite, but I would recommend specifically that you know linked lists well as this course heavily builds upon the concept.

A very important course for anyone doing computer science or software engineering, since the content is quite relevant for the nature of and types of questions you may see in programming interviews when applying for internships and jobs. This course will also enable you to write better code, since you will now have enough knowledge of algorithms to select the appropriate one and analyse how efficient the programs you do write are.

Lab work is quite straightforward and less in the character of COMP1511 with many short tasks, instead preferring to have one or two more substantial, related tasks on the main programming topics. The assignments can be time consuming in figuring out what you need to do and implementing it, so definitely don’t sleep when it comes to starting on them. The midterm exam is quite reasonable - if you’re confident in your programming, it can take as little as a few minutes to complete. The final exam is a straightforward test of your theory knowledge of all the topics in the course with a programming section that will usually focus on the main programming topics - linked lists, trees and graphs. When it comes to preparing for the practical section, I definitely recommend trying some of the problems on LeetCode et al.

In the event that your exam mark was lower than your assignment marks, your assignment marks were adjusted by taking their harmonic mean. This scaling is always regressive, meaning that your mark would never be adjusted up, only down. There are reasons to use this - a big one is to punish plagiarism - however it only really sought to punish people even if they didn’t cheat. However in 19T3, this was changed to only affect students for which there was a 45% difference or more in their assignment and exam marks. I can almost guarantee that because of this change, course marks were higher on average this term than they have ever been. I bring this up since in future offerings this may be tweaked yet again, so your mileage may vary.

- 2x 2 hour lectures
- 1x 1 hour tutorial
- 1x 2 hour lab class (follows the tutorial for a combined 3 hour block)

2/5

Yes, screen and voice.

Dr. Ashesh Mahidadia

Lecture slides with some extra programming exercises - this obviously depends on the offering since earlier terms reported no such exercises, but Ashesh should release these in subsequent offerings from now on.

4.5/5

None prescribed, however the recommended book is “Algorithms in C” by Robert Sedgewick, published by Addison-Wesley.

19T3

99 HD

Prerequisites:

An interesting course, with a few difficult parts, and very useful overall. One of my gripes with the course is the harmonic scaling applied to assignments and final exam results. If you get a lower mark in the final exam than your assignments, your assignment marks are scaled down using the harmonic mean. It's used to stop plagiarism, but serves to punish students regardless, since you're almost certainly going to go better in the assignments than the final exam. Didn't attend lectures so I can't comment on the lecturer, but the written materials were more than sufficient, and the tutorials went along with them well. In week 4 there was a "Sort Detective" lab where you were given two programs (compiled and with no read permissions so you couldn't decompile them) and had to run tests on them to figure out which sort they were, and write up a report about it (like a couple pages or something, not a proper report), which I found interesting.

2hr + 1hr Lecture, 1x 1hr Tutorial + 2hr Lab (Combined)

2/5

Yes - screen and voice recorded.

Dr. Ashesh Mahidadia

Lecture slides and course material uploaded to WebCMS.

3/5

Note: I don't use textbooks and can't comment on their usefulness

2017 S2

96 HD

A bit complicated. Involves labs, assignments, prac exams and final exam.

COMP1511 is the sole prerequisite and is sufficient for this course.

This course is one of the immediate continuations of COMP1511. It replaced COMP1927. UNSW CSE recommends taking it in second year, however there really is no problem taking it in the second semester of first year. It is a prerequisite to almost all future computer science courses.

The disappointing amount of marks I threw away in the final exam killed a bit for me. It's really my own fault and the course ain't to blame for it, but it does skew my rating quite far away from what it should've been.

Algorithms are perhaps the most fundamental tool computer science students use. The study of algorithms involves trying to solve problems as efficiently as possible and also create the program to do so. On one hand, you need to understand how the algorithm works, but then on the other hand you need to know (or figure out) how to implement it. They're quite fun to explore when you sit down and attempt to understand it.

This course serves as the introduction to algorithms, with two areas of focus: sorting and graphs. Sorting is quite self explanatory, but I think this video helps introduce the fun behind it all. The graph structure forms the basis for shit tons of stuff we do (it becomes clear that even Facebook is essentially just a huge graph). It leaves you with pretty much all the basic needs for you to get a job as a computer scientist, and you really should never forget what you learn in this course. It's just that important.

At times, I found some of the lab exercises and assignments quite draining. The only thing that bugged me was that occasionally, our tasks weren't clear enough with what we had to do.

I feel like the rumours that maths helps computer science starts here. My mathematical background certainly simplified a lot of bizarre algorithms for me a lot. But I don't think they're necessary though. At the end of the day, the algorithms are backed by logic, and that's a skill any computer scientist must have.

3 hours of lecture, 1 hour tutorial followed by 2 hours laboratory

2/5

Yes

Dr. Ashesh Mahidadia

Lecture notes, lab exercises, tutorial exercises all uploaded to webcms3.

3.5/5

- Algorithms in C, Parts 1-4: Fundamentals, Data Structures, Sorting, Searching (3rd Edition) by Robert Sedgewick, Addison-Wesley
- Algorithms in C, Part 5: Graph Algorithms (3rd Edition) by Robert Sedgewick, Addison Wesley
Both are VERY good, but not needed for this course.

18 s1

85 HD