Data Mining and Statistical Learning

As others have pointed out, the lectures feel like a waste of time. The quizzes are also not that useful and closely resemble (and often directly replicate) the practice questions, so as long as you've done those beforehand, you should manage fine.

I’m giving a neutral rating mainly because I found the homework and project enjoyable, which balance out the more mundane aspects of the course. The flexibility to choose your own topic and dataset makes the project engaging, and I got to follow my peers' train of thought through peer reviews (yup, there are peer reviews). Overall, the course builds reasonably well in terms of content and expectations, which is a decent follow-up to ISYE6501.

Positives: -It's super easy, relative to other classes I've taken. That might be a good thing. -Great freedom in the assignments: basically do what you want and write your report. Very free form and exploratory. That was nice. Very business oriented in that way. -No mandatory group work! Group size of 1 is allowed.

Negatives: -It's super easy. As a data science hiring manager, this course makes me doubt the value of seeing GA Tech OMSA on someone's resume. The quality of reports I peer review are so bad. It is easy to get an A in this class without really knowing the technical details in any depth. -The lecture videos are just the prof non-stop rambling through equations over and over and over. It's really unengaging and detached from the assignments. I stopped watching them.

There were many weird, bizarre things in this class. A few examples:

• HW2 dataset has brozek as response, but siri and density as predictors? They all have ~0.99 correlation. What a bizarre assignment.
• A TA told me not use to emojis in my reports. Not only were the assignments boring, but then the TA sucked the fun out of them in grading.
• A lecture slide says "accept the null hypothesis if p-value is large." Come on, you never accept the null!
• Inference of modeled p-values and confidence intervals without any prerequisite discussion of model assumptions. This trains people to hit model.fit() and automatically believe the imported library's print out. Really bad.

In all, this class was what I expected: an easy 3 credits. Ultimately I regret taking it because it was so boring I could barely muster the motivation to do the (easy) assignments.

I am an OMSA student, and this is the last course in the program for me prior to practicum. Anyway, this class was perfect for me, given where I am and what I need.

I came into the class with some confidence in data manipulation in R, and I am leaving with extreme confidence in the same (as I sidestepped base R data work and committed to using tidyverse), and also in using a suite of ML models.

The latter is kind of what I wanted out of the class – a place to actually build and tune models, to experiment with cross-validation and data preprocessing, to learn some exploratory data analysis techniques. We did this for both regression and classification tasks across five different homeworks plus a final project (I did solo), and at this point I feel VERY well-equipped to actually build out my practicum models. Each module comes with a set of commented sample code for the ML methods in the module – these and the walkthrough videos are great for learning how to actually ‘do’ ML.

Does the class have downsides? Sure – the lectures are not great, and yes you have to watch them or at least read transcripts to do the knowledge checks so you can get full marks on quizzes. Yes, the lectures are theory-heavy and no, the quizzes do not really test your knowledge of theory. I won’t say the quizzes are useless though – you have ample time to learn what you need, and much of it does stick.

Strong recommendation if you are anything like me.

I really wanted to love this class and hoped to get 2 things out of this class. 1. to gain a better understanding of the stuff I learned in 6501 (both theory and how to apply) 2. how to use real dataset to model in R. This class was so far below my expectation that I would completely recommend people not to take it, unless you just want to take a relatively easy class.

Lecture: I watched the first 4 weeks, and then realized he just reads the slides, which are terrible to begin with. Lots of notations not explained. It typically goes like this: Starts with - Some high level explanation of the model ("logistic regression is a supervised model used for classification", and then immediately: 7 pages of poorly explained math formulas. But not only poorly explained, but no context, or explanation of assumptions. I ended up watching stat quest and just googling other materials to understand it better.

Homeworks: They completely ignore the problem with real life data sets, that you'd have missings, formatting issues, and other problems that you have to clean up. Instead you are given super clean datasets and you just need to tweak R codes already given. Not only that but some homeworks are super repetitive and long, where all you're doing is changing 1 parameter over and over again. I can say I learned next to nothing from the homeworks.

I also attended a few office hour sessions where TAs are not good at explaining basic concepts, the quizzes are easy to score well on, but also don't make you learn anything fundamental about the material, it's just cherry picking a few niche areas from the weekly knowledge checks. This is like the typical class I experienced in undergrad where you go through a bunch of hoops, and at the end, learn nothing that can be applied to real world.

The concept of the course is fun. Lots of different modeling techniques, take the dataset and run with it, write a report.

The execution of the lectures was horrible. I ended up not watching them because the professor just read off the page, and the content wasn't useful anyway. It was more useful to research the packages we were using on my own.

What really sucked was peer grading. All over the place. I put minimal effort into some reports and got a 5/5, put a ton of effort into others and really knocked my model out of the park and got a 3/5 with relatively no feedback. 30% of the grade should not be peer graded - a major failure point in this course and most courses in the OMS programs.

I loved this course. I considered this to be a great sequel to ISyE 6501, as it delves deeper into a range of widely used modeling techniques. Like many others have noted, there is a stronger focus on writing reports and applications. I found this to be very useful, as communication is one of the most important skills for a Data Scientist.

The group project was a great opportunity to apply what you've learned with any dataset of your choice. I loved the lack of exams - instead there were 5 quizzes, and a 'take home' final, which was more of a 'solo project'.

This course covers some material covered in CDA. This course is a great alternative if you are more interested in applications rather than coding ML algorithms from scratch, but both complement each other as well. One of the common critiques of this course is that the starter code for the 5 HW assignments is in R. However, students are free to use other languages such as Python, which I used for some of the assignments.

Overall, this was my favorite course I've taken in OMSA. It was not overly challenging - it felt that my efforts went directly into truly learning. The professor is very responsive and I enjoyed his positivity and teaching philosophy.

One of the better classes I've taken in this program. The lectures are pretty much useless, but the homework and final are fair and kind of fun. You basically get a dataset, a problem, and some starter code (less code as the semester goes on) and that's it. You write your code using the models instructed, then write a paper on what you did and what you found. I enjoyed most of the assignments. There's also a project, which is similar to the homework, but you find your own dataset and use whichever models you want.

I took this during my last semester. Probably one of my two 2-3 courses. The material and slides were very well organized, the teacher was extremally active and hosted office hours each week (and would test students on what they knew during office hours--which I thought was great), the TA's were kind, knowledgeable, and helpful. From a material and learning standpoint, while not as hard as CDA, still very valuable IMO. The general concepts of ML were done over and over again and I felt like it was a great opportunity to do a ton of practice on different data and models. It really drilled the concepts into my head. Also, technical summary is really important in this course. In the real world, if you can't summarize your work well, you are out of luck. It is a vital skill and I appreciated that aspect of the course. I will take the general structure with me. Definitely worth taking!