Hands-on with Applied Analytics (IS5126), NUS, Jan-May 2025

• Great teacher, really increased my understanding of ML methods and how to apply them. Covered alot of groundwork in this course that I'm sure I will use in the future.
• Very nice teacher!
• The teacher is highly dedicated to teaching and consistently approaches each class with a strong sense of responsibility. He actively encourages student participation and engagement. Through his course, I not only gained substantial knowledge but also broadened my perspective, making the experience truly rewarding.
• The principle is explained in great detail.
• Best ML teacher I've met, I learned very useful skills, and he know all these knowledge very clearly and can teach them well.
• very passionate
• For providing very structured course and dedicated way to explain concepts each class.
• Very passionate about helping students understand the concepts in class. I also think that the course was helpful in deepening my conceptual understanding for industry!
• Professor helped me get started in the field of machine learning. The course content is hardcore but helps us understand the essence.
• One of the best prof in NUS
• good teaching style
• Waits until everyone is clear and doesn't rush. Explains topics very well
• Nice person
• His lecture is so interesting and I am encourage to study hard.
• He is committed and pushes students to do their best.
• Beneficial for deepening understanding of model scenarios, math, and code. Amazing.
• Excellent digging out the essence and really care about students
• I think Prof. Kok works very seriously.
• nice teacher!
• The lectures are very good.
• Great approach to teaching, we really delved into the guts of alot of the machine learning methods. I really like that.
• In alot of other courses we just load a package and run a couple of commands, but here we built our own and then applied them.
• Excellent in any aspects.
• Math and machine learning
• Gives well structured course and the flipped classroom method ensures that we come prepared to class. Puts in lot of effort to teach all concepts in a clear manner so that students from all backgrounds can understand.
• Very passionate and focused on helping students deepen their understanding of the material!
• Very passionate and makes the lectures engaging
• Solid knowledge, Quick Feedback, Commitment, Plenty of Content
• memorize the students' name, strong math knowledge. Good understanding of students' logic/questions, hence his explanationn is easy to digest.
• Takes a lot of effort in giving in-depth analysis
• His vast knowledge in the Machine Learning Mathematics and concepts and knowledge.
• Well prepared for the mathematics theory and the derivation is absolutely fluent and smooth.
• Excellent at digging out the essence and math prove of all the concept, and really care about whether we can understand.
• I think Prof. Kok works very seriously.
• Theory
• He demonstrates strong expertise in the theoretical foundations of machine learning. I appreciate his ability to present mathematical derivations and proofs of the algorithms taught in class.
• Maybe upload the notes so at least we can compare what we have written down during class.
• I like that we all have to come to the lecture every week so maybe there's some way you can distribute the notes to only those that come?
• It would be useful if the classes could be recorded
• NIL
• Workload so large
• The labs seems a bit too old school. I understand the value of implementing neural network from scratch, thou numpy seems reasonable enough to be allowed. If this was actually allowed, better to be communicated or imported into the lab files.
• The module name 'Hands-on machine learning' is a bit misleading. I though 'hands-on' here is about exploration of latest practical ML techniques and libraries. The hands-on here seems referring more on using pen and paper to take note and derive mathematical proofs.
• NA
• Do not set so hard exam paper please.
• I think the first video is too difficult , easy to discourage beginner Besides, there are so many formulas in class that it is not easy to remember all , even the top significant formulars,, Still it's a little difficult to deal with the codes if we are not familiar with scikit-learn, pandas ,,etc before.
• Slightly less prove maybe.
• I personally feel that the final exam could carry more weight in the overall assessment.
• Please reduce the compulsory video learning type and theory revision before quiz not ask to answer directly at the beginning of the class at 6:30 pm. Kindly considerate for part time students who need time to reach class from work.
• While understanding the theoretical derivation of algorithms is important, I feel the course could benefit from a stronger balance between theory and practical application. Given that the course is titled "Hands-on with Applied Analytics", a heavy emphasis on mathematical derivations may not be the most relevant for students aspiring to careers in data science. For example, testing students on memorizing the prof of specific parts of the Value Iteration Algorithm which taught in class has actually little to do with their understandings on the field.
• In contrast, insights into how data science teams operate in real-world settings, including their infrastructure and division of labor, would be more applicable and valuable. These topics were more comprehensively addressed in the flipped classroom videos by Geoff Hulten, but were not adequately covered in the live lectures. Overall, I believe the course would be more effective if the live sessions also incorporated practical perspectives, offering a more balanced approach between theoretical understanding and applied knowledge.
• Give us a glimpse what's exactly behind those ML models, and what's those concepts are in mathematical way.
• The course gave a solid understanding of the theory behind machine learning algorithms. It was clear that the Prof knew the material well, and his detailed derivations helped me understand how the algorithms actually work under the hood.
• This course offers a systematic introduction to the principles of various algorithms in machine learning.
• give me a overall intro about machine learning.
• Usual class
• Wide range of topics
• A very broad course of ML
• Went into the guts of the ML methods. Didn't just load a package and run some commands.
• Useful knowledge
• Flipped classroom, the prof, the course structure
• strengthen my understanding on ML
• Very detailed
• The hands on aspect.
• a bit more prove so that some may not be clearly understand
• The course felt too focused on theoretical derivations, which didn't quite match the "Hands-on with Applied Analytics" title. There was very little discussion about how data analystic is done in practice -- things like team roles, project workflows, or infrastructure. These were only mentioned in Geoff Hulten's videos and not really covered in the live lectures. I think a better balance between theory and practical insights would've made the course more useful overall.
• The final project takes up too much of the total grade and the workload is quite heavy. I think it would be more reasonable to remove the guided project and replace the first assignment with the content of the guided project.
• Too much complex math problem in the class. This is not as same as course guide.
• NO subtitles,, and some Q&A from students , the voice is too slow,,
• None so far
• Final Project Feedback was not communicated effectively earlier
• NA
• Big Workload
• too much topics covered in surface level and lacking practical aspect (coding / design) during lecture
• Huge online video learning time is demanded
• Difficulty of exam.