Kursteilnehmer-Bewertung und -Feedback für Introduction to Recommender Systems: Non-Personalized and Content-Based von University of Minnesota

Well prepared course. In-depth lecture. Easy to follow even when listening only. The course lectures is very detailed, and that is one thing I really liked. The videos does feel a bit long, and maybe we can chop it to smaller sub-topics.

The interviews are very interesting and show a glimpse of broader universe of recommendation system. However, the concepts explained in the interview is a bit hard to follow, as there is no accompanying presentation materials and it jumps to detailed content with little context

The regular exercise feels very easy but helpful to make the concepts concrete. The Honors programming exercise looks interesting & challenging, but it seems too hard for someone with no programming background. I am also learning Python in parallel, so I decided to drop it to avoid learning 2 languages in parallel.

videos are too long... I had to watch them with x2 speed...

Too theoretical. I hope other parts will have more details.

Very informative course! I had a great learning experience working on the programming assignments required for honors. The only drawback is the style of communication (written and spoken) is elaborate and confuses many non-native English speakers including me.

I am very new to recommendation system and yet able to comprehend the lessons. The best thing is explaining the system with example. Walking through Amazon.com and explaining content based and collaborative filtering is easy to grasp.

I think this is a good course to start exploring recommendation systems.

The course was easy to understand. but i find the slides not much of help.

first time taking a course using Coursera...material was very interesting and well explained. I wish there was a way to speed up the audio track a little to shorten the lecture length. hard for the lecturer to engage with an audience that is not there, but both tried to do so.

I think some of the interviews didn't really give me great insights. I know this is only an introduction, but I was expecting more fields than movies. I am overly critical though, all in all a very good way to understand recommendation systems.

It would be better to make practice on Python.

Great introduction to Recommender systems. Really got me thinking about how I could apply them.

Some of the assignments are not particularly well created, in the sense that they seem to emphasize on recalling rather than learning, Also, most of the interview failed to hold my attention in general.

Overall, however, this is a very good course and gives a comprehensive overview of the prevalent techniques in the relevant fields.

Overall a good course that teaches the basics for content based recommenders.

Would be great if the assignments were a bit more challenging, e.g.: work with large datasets (and not the tiny datasets used in the assignments)

Would also be good if we were provided papers of recent/notable research on the topic to read further.

The course was a good one with content that's understandable. I can't wait to proceed to the next one

The lecturer were very lengthy, at least for me. I find it difficult to concentrate.

The content of this course is solid. It's a good introduction to content based and non-personailzed recommender systems. However, the presentation is poor. The course is largely based around videos which appear to be single takes. Snappier, well edited videos would have been better and, as a result, I often found myself skimming the transcripts rather than watching the videos.

I expected a lot from this course but it could have been a lot better - lengthy videos, not trying to explain the concepts in an understandable ways. Ended up confusing with various interviews and what are differences between various content based recommenders. The programming exercises were good and provided a good overview.

As an introductory course, the content was good. But I wish the approach was more analytical and more hands on. Rather than history of Recommender systems & what happened in the 90s, I would have been happier if the course was able to throw light on the latest stuff in this field, the latest mathematical techniques etc.

Pros:

Some useful terminology if you want to ever communicate with someone who does recommender systems.

Cons:

Very diluted content.

Mostly large text slides with the presenter talking in a monotone voice.

Programming exercises are done in Java and require deploying an IDE + an unused open source project developed by the authors. Hint to the authors: use Python, R or Octave like everyone does.

Some of the questionaries are ambiguous.

Generally I like the contents of this course. I particularly like that insights are provided in terms of what aspects to consider when designing a recommender system; pros and cons of different approaches. However I'm also extremely bored watching the videos because looking at the lectures reading the scripts (most of the time with very slow speed) is one of the quickest way to send people to sleep. I'd hope the lectures will improve their presenting skills.

Another comment is the honours track assignments should really be put into more thoughts. I passed them with 100% credit, but I didn't feel I gained a lot useful knowledge through this exercise. Generally it felt to me that the complexity of the implementation is much much more than needed in relation to the complexity of the problems. Eventually this assignment became grinding with Java's verbose, annoying syntax and unnecessary computations designed in lab instruction. For example, in the first programming assignment, why if the ModelProvider object already computed the entire map of ratings, and the map is directly needed in the Recommender object, the Model object only provide API to retrieve individual rating but not the entire map?! Isn't it a wasteful computation to reconstruct the rating map? So I doubt the structural design of the program is sensible, or the expected solution would actually be done in real applications. Also I think Java is just a really out-dated, bulky language to work with in this kind of task. It really makes the assignment experience awful.

The assignments are not very relevant to what is being taught. Java 7 instead of Java 8 makes things too verbose. Lenskit is painful to use and in the week 4 Honors assignment its just impossible to get the results desired by the grader. I would suggest the Teaching team to use R/python scikit instead of Java

Ok, it's an introduction, but it could at least show us some math or pseudocodes. A part from that, the course is really awesome. Well structured classes, good explanations and incredible interviews

The introduction is very slow in my opinion. Hopefully, it will pick up the pace in the later modules.

Not bad for an introduction, but I would have prefered it to be more technical

It would be nice to have a hierarchical overview of the recommender systems. It's easy to get lost which is a subcategory of which. Thanks for the course!