Building Recommender Systems with Machine Learning and AI (Udemy.com)

I would give this course 10 stars if it was possible! It's not easy by any means, especially if you want to understand / run /modify scripts, but it' totally worth time and effort. Packed with valuable information, sometimes too condensed, but it's up to you to explore it further. Besides, Frank - the instructor - is super helpful, responds to questions literally within a few hours. Thank you, Frank!

Thanks Frank for making this great course with lots of information and useful code samples! Students get a very deep insight into which systems there are. Of course, due to the sum of topics, some are only superficially treated. I would have wished for a little more theory lessons to explain certain concepts, like the matrix factorization or the "Embedded-layers" in the TFRS. For some topics it's not clear how dated the information is, since Amazon DSSTNE is officially "archived" and hasn't been updated for at least 3 years. I really recommend taking this class!

I have an basic understanding of recommender systems, that said, I am newbie from a technical standpoint and various algorithms that is basically used for recommendations. This course provides an excellent insight on various algorithms and goes little deeper on some of them. Few algorithms and terminologies like SVD, KNN, RNN , soft max ,classification and segmentation, binary vs multiple classifications are explained at right level to get an overall understanding of how these can be used in recommender systems. Overall a Great course. Got a very good insight and would love to revisit this course multiple times and try all examples explained. Great course indeed .

Frank and the Sundog team delivers some of the best quality in the area of recommendation engines and big data applications; having developed some of these at Amazon and IMDB. This is a great survey course to get you caught up on everything, and learn the latest across a wide variety of algorithms and platforms. The deep learning section was really a nice surprise with it's depth; I have had several courses covering deep learning and this taught me some more new things. There are lots of papers/resources available and the code examples are excellent to learn from.

Definitely worth it. You will really enjoy this course, but you need some background in OOP with Python. Even though all code presented here is carefully explained, you must take your time to understand the nuances. The instructor guides you throughout the entire course and the course itself is well-structured. The bleeding-edge alerts constitute an important part of the education process here. Now I feel ready to start reading more advance stuff. The instructor gives you many paths you can follow after taken this course.

This a practical course for what is involved in building a recommender system. The course covers a lot of algorithms, explains the theory and provides a lot of code to get going with. Doing the examples requires some coding and a bit of time to write, debug and run (especially if you're using a regular PC). Frank Cane focuses most on what works in real life, and distinguishes between interesting techniques that are unproven and those that have been battle hardened.

Very in depth course into the recommendations world. I liked how he explored a lot of possible ways to solve the same problem. The coding can get a bit heavy to understand in some points, but overall the main take home concepts were very well explained and concise. This has proven to be an invaluable resource during my former job as a senior data scientist where I had to design a rec system for a ecommerce site. This course only is much better than 99.9% of anything you will find online. Frank really knows his stuff. Thank you so much Frank!!

Very good course, but the student will need to do a lot of off-the-clock research (read papers, single step through code looking at variables not explained) to apply the high level material in the course. There is also a lot of Python magic used as well as matrix operations that are used not well explained. None the less, the class is a bargain with unique industry experience that is worth its weight in gold.

This course is very useful for practitioners in recommender system field. I would not start my tasks without this course. But you need to spend much time on studying source code because the information in the lecture is very limited. Frank is an expert on recommend system with rich industry experience. He give many good advises on real word recommend systems. It should be better if the course material can be designed more elaborately.

The course is quite good as far as the intent and spectrum is concerned, but it seems t be packing too many things in a single course. As a result, the due time given to each concept takes a hit. Important concepts such as these cant just be 'spoken about' as if they are living room small-talk. There is an unsaid urgency to rush through everything and that puts the burden on the student to self-imagine a lot about the concept which is being just spoken about. I took every exercise seriously, followed and solved each one of those and frankly they are quite good. However, more explanation, more visuals and spending more time for each concept will go a long way in ensuring that students stay engaged and not struggle to keep pace with jet speed delivery.

Very good course.

The instructor is very clear in his explanation. The course is structured well and it's not hard to follow