Before getting into the course evaluation: James seems like a well-intentioned instructor and good person who put a lot of work into this course. As a former teacher myself, I know it can be hard to not take criticism personally. So please know that none of what is below is personal.

The course is a series of linear algebra drills with no connection to data science concepts or applications. A data scientist needs to develop a conceptual foundation that helps them **use linear algebra to solve data analysis problems** using programming languages or statistical packages. This is a lofty learning outcome and a big shift away from traditional math education practice, so it would be understandable if a course were to aim for this and come up short.

This course, however, is simply drill and kill math with no clear objective. There was no connection to data science, no demonstration of practical appliations, no focus on higher-level conceptual understandings. Memorizing procedures that in practice are always carried out by computers is only valuable to the people who either will will write code to implement the procedures from scratch or will develop new mathematical procedures entirely. I'm guessing neither of these groups are not the target for this course.

An analogy: Imagine somebody offered a "Thermodynamics for HVAC Technicians" course and then spent weeks on hand calculations of energy transfer or entropy in the abstract. Being able to calculate energy transfer by hand isn't necessary for learning to fix an air conditioner, and being able to calculate eigenvectors by hand isn't necessary for learning to use PCA. The concepts are the point: What data science problem does an eigenvector solve? Why does the eigenvector solve it? How can one make sure they're using the eigenvector effectively in real-world context (in which the eigenvector has been calculated by a computer). As a practicing data scientist who never took a formal linear algebra course before, I can say that I do not know the answer to those questions any more so than before I took this course. That is my biggest disappointment.

A more effective course would start with a data analysis problem, then walk through conceptually what solving it requires, then introduce just enough linear algebra concepts to help the student develop that "under the hood" understanding that allows them to recognize when to apply the solution to other problems and how to do so effectively. Quizzes would then give them an opportunity to do just that.

Perhaps subsequent courses in the series revisit the procedures drilled here and eventually illustrate their application, but it's the application that should be the focus up front, with linear algebra concepts introduced as needed to solve the problems. Hand calculations would be used sparingly if at all just to build conceptual-level understandings. Courses would build up with progressively more challenging data science problems to solve, with each introducing the relevant linear algebra concepts just-in-time to solve the problem.

[Really, this course deserves a slightly higher rating, but read to end about why I gave it a 1/5]

The good: James is clear and his penmanship is good.

The bad: I've been working in the data science field for several years now, and I felt that it would be good to refresh my basic math skills to get a deeper understanding of how certain methods worked. I was excited to take this class simply for the sake of learning, I don't care about certification.

To my disappointment, this class designed "for Data Science" (it's right there in the title) didn't explain ANYTHING in terms of data science. In fact, it didn't explain anything at all. To see what I mean, go to the lecture on eigenvalues. The introduction was basically, "now we are going to calculate eigenvalues, here's how you do it." Excuse me? Eigen-what? Why on Earth would I want to know about these things? Can you please give me an example of how they are use in the real world? [Aside, I often kept ChatGPT open on the side so I could ask these kind of questions, but it shouldn't be necessary.]

Truly, the lack of any relevant context was absurd. Could this course at least imagine that I was somewhat familiar with the idea of regression and teach me some things with at least that much context? Sure, this might be a fundamentals class, but by the time anyone would actually get to the real data science (where they would be using a computer to do these calculations), all of this material would be forgotten because it was taught in such an abstract, meaningless way. [Another aside, unfortunately, this is all too commonplace in mathematics courses, but it's not an excuse.]

Look, I'm sure James is a nice guy, and perhaps a good teacher in a real world context. But this content should be embarrassing to him. I'm going to give him the benefit of the doubt and say that the issues were probably administration related (poor incentives, mandated curriculum), but I wish someone had stepped in and said, "hey, let's actually make a good class where students really learn what's under the hood." But maybe that's asking too much for a massive online course.

Moreover, it seems pretty clear that nobody maintains anything here or checks the discussion board. Someone brought up an issue with a quiz question being wrong a year ago, and it still isn't fixed. But, most students likely don't care because if your "test" is a single multiple choice question that can be retaken as many times as you like, it hardly feels worth the effort to demand any changes.

So, in the end, like I said at the top, I gave this course 1 star because I want someone, ANYONE, to actually care about this. Shoot, I'll even go up to 2 stars if someone on staff simply responds to this review. Going up to three stars would require some updates to content though, so I'm skeptical.

In the end, I'm just disappointed. I came here to become a better data scientist. For the moment I can calculate the inverse of a matrix, but I don't have any sense of why I would want to - and I will forget within days. If you're here for a certificate go ahead and plow through this, you can probably do it split screen with something more interesting. If you are here to learn, don't waste your time.

This is a very well explained course, put together nicely with bitesize lessons that consistently keep the dopamine flowing. An added bonus to this course is the topic area, in relation to the under-subscribed Tech industry, giving learners an added edge when attempting to enter the Data Science field.

Professors breeze rough concepts without any explanation as to what is going on. This course is supposed to be a bridge course that allows students with limited exposure to learn basics so they can begin a data science learning path, not a review course for people who have already taken the subject.

For example, professor does NOT explain what a determinant is (extremely important concept) but spends time teaching how to solve determinants by hand. Memorizing formulas without any reference to what is actually happening is a useless practice and not suitable exposure for students attempting to learn a new subject. The entire specialization follows this format. feels extremely half baked.

A very succint course. It's great for a beginner and can be easily understood. :) The quizzes are easy too.

Simple to follow. Easily explained. Good for revision as well as starting fresh

Great introduction to Linear Algebra for beginners!

A quite succint course. Good for review or have some basic understanding of linear algebra. But if you want to have a thorough understanding of linear algebra, I think this course would not be enough and you should maybe use this course as a supplementary.

Good side is that the course discusses many essential topics. But basically that's it, the course does not flow smoothly, the design of this course is not very good, and I found the discussion on eigen values and eigen vectors are too basic and there are some confusions in quiz questions/answers. I don't recommend taking this course no matter if you have not learner Linear Algebra before and want to get a start, or if you have and just want to get a quick review of it.

Well-explained and comprehensive. I thought it was going to be a rough course but Professor Bird is very thorough and concise in his lectures.

Very well done, conducted at just the right pace

One of the best courses that I have found in Coursera. The material was challenging, but not overwhelming. Additionally, the istructor was phenomenal. My one recommendation would be to also go a little bit into how the computations are applicable to data science, rather than just go through the mechanics. Applications would likely be better in a follow on course, once the student is confident with the basic mechanics. Nevertheless, a truly phenomenal course and instructor.

I love the way the course was conducted with fluency and the quizzes are no problem although not a sail through either for someone who has attempted multiple external questions.Finally, it is good for a beginner-intermediate individual or anyone looking to recap their L A concepts.

Definitely not an introductory course, but a great high-level overview of LA. If you are new to LA, I highly recommend watching the 3Blue1Brown LA series on YouTube followed by Khan Academy's LA series. THEN tackle this course. Overall though, it was a great delivery and truly is the essential LA concepts needed for Data Science.

Course covers basic linear algebra calculations and formulas, but does not provide intuitive understandings of what each concept entails. Very plug-and-chug heavy course where you have to memorize formulas presented in lecture and act as a human computer on graded assignments. If you want a great supplementary resource, check out 3Blue1Brown's video playlist on Youtube: Essence of linear algebra - at least there, you can gain an intuitive understanding of what all these calculations are doing.

The course is to-the-point revision of essential Linear Algebra for Data Science, Supervised Learning and Business Analytics. It requires some intermediate-level background in Mathematics for the assignments of the last modules (weeks). It gives an outstanding mathematics analysis for a further algorithmic-based courses in Data Science by the essential mechanism of vectors/arrays.

I thought this was a really great course. The instructor explained things well and didn't assume that you know everything already (like some other courses). This wasn't even a required course for my MSDS program but I found it very useful for filling in some knowledge gaps and will help me as I continue my Master's in Data Science.

An awesome introductory level course to linear algebra. Very succinct, well-organized, and covers most of the important basics of linear algebra. Definitely recommend it to students who are completely new to linear algebra or students who have some previous knowledge about linear algebra but need a review.

Decent intro course, but not a lot of exercises. The strategies are slightly different from other courses. About the same difficulty as the deeplearning.ai intro to DS Math, much easier than the UCL one.

This course is easy to understand for a non-native speaker like me. I learned a lot from it and it encouraged me to build my confidence in mathematics. I am grateful to all the teachers of this course.