Great thanks to this amazing course! I learned a lot on state-to-art natural language processing techniques! Really like your awesome programming assignments! See you HSE guys in next class!
It's a comprehensive course on NLP. The instructors clearly explain both the traditional/classical approaches and modern approaches such as neural networks in NLP.
автор: Bayram K•
I can't see anything educational in this course. I frustrated. Simple recitation of facts from NLP articles and nothing more. Programming projects are disconnected from videos., i.e, upon watching video don't expect to understand how to complete a programming project or at least to understand what you have to do in order to pass the assignment. This course is the worst I have ever seen on Coursera.
автор: Daniel I G•
The course is very poorly explained, in comparison with other similar ones in Coursera.
Unless you have already some knowledge on the matter, it's pretty difficult to learn it just by watching the course's videos.
автор: J A•
Very good advanced course on NLP covering both statistical and deep learning modeling approaches. Very well cemented by programming assignments in python and real-world acht bot application using Telegram and Amazon Web Services free tier solution. Check the pre-requisites because this is an advanced course requiring some previous knowledge of NLP, linguistics, machine learning, deep learning and above all python programming and some AWS and Docker basic knowledge. Staff is very responsive and supportive on the forum. Finally, don't underestimate the time it takes to go through all videos, quizzes, and programming assignments -the guidance given is on the low end.
автор: Hanna P•
A lot of information on a very high level but with reference to papers. Papers are a little bit dated for now, but this is due to rapid progress of NLP as a field of study. Assignments which need to be peer reviewed are painful, because you don't know when someone will check your work. This is due to not many learners simultaneously at class. In general, the course provides helpful advice when working with NLP and reveals some of the core approaches. Also, there is enough practice here, which is great.
автор: Aleksandar J•
Terrible final project realization. You use telegram to deploy your conversational bot, which shuts you off if you don't have people chatting with it for too long. And that can happen if nobody reviews your project for days and weeks. I give this course a one star, as this kind of thing is just unacceptable from my perspective, and I can elaborate why. You've taken my money for waiting more than a month for reviews, only to find out I did not pass because my bot shut down. I'm sorry, but I won't take another course from you again. Goodbye.
автор: Alexander S•
This is the most rigorous NLP course that I could find on Coursera or any other MOOC provider. A thorough introduction to both classical and NN models for a variety of NLP problems and tasks, drawing upon recently published articles.
The quizzes and programming exercises do at times exceed the content covered in the lectures, but if one does not mind supplementing the lecture content with self-study/problem solving, then this is not so much an issue.
The main suggestion that I would give is to reduce the numbe of peer-graded assignments (there were 3 in the span of this 5-week course). The peer grading component is time-consuming and error-prone.
автор: Xinghao G•
Are you kidding me? What are you talking about in courses? I have to re-study everything after each video because it just doesn't help you understand the theories at all.
автор: Tianqi L•
Anna is a great instructor. She can explain the concept and mathematical formulas in a clear way. The design of assignment is both interesting and practical.
автор: Vasudha A•
The course instructors do a bad job of explaining the basic concepts for anything in the Week 1 itself. They are inclined towards using jargons form the field, providing no context of basis of using the different algorithms or concepts. This course is probably, only, suitable for people who need a quick refresher on something they already know very well. I will not recommend this course if your aim is to LEARN NLP.
автор: Umnov A•
The course had a very good content, but it's made very poorly. Lectures are uninformative and feel rushed, there're virtually no reading material which would be useful here, parts of code are broken by now, noone tend to community forums so if you have a problem that noone encountered before, you are all by yourself. One more thing about lectures is that they don't feel prepared at all, it looks like couple more takes and some rehearsal would make a big difference. I'd recommend spending money elsewhere and kinda regret doing that myself.
автор: Marcin G•
This is an advanced course. Creators assume you have significant amount of knowledge in the field, if not you have to do additional research. The course covers most up-to-date achievements in the field of NLP. It covers widely used techniques of improving NLP performance. You get a chance to see interesting point of view from some of the best people in NLP. Well prepared and thought through material. Teachers are enthusiastic about the filed, which helps learning. Definitely worth taking if you practice language analysis.
автор: Ignacio G•
Shallow theoretical explanations (just mentions and references). Disconnected assignments from theory.
автор: Samuel Y•
The course covers traditional statistical techniques as well as some latest deep network methods, and gladly both the instructors speak good English on the course of Natural Language Processing :-) . The assignments of this course are also more guided and easier to follow than the previous courses of the specialization, meanwhile it will require quite a bit of environment setup efforts like docker and server to finish some assignments. Having access to some GPU machine/environment is recommended to speed up the training/testing process.
автор: Qingsheng L•
Course material is not up-to-date: You cannot imagine an NLP course that doesn't know Transformer or Bert in 2019.
Assignments are too easy: It isn't really useful for understanding the algorithms introduced in the course.
Not deep enough: Many algorithms were introduced, but briefly. It doesn't help you prepare a interview or anything, because without a deep understanding in the algorithm, you can hardly impress the interviewer.
In general, I don't recommend this course. It's not hard to accomplish and earn a certificate; but it can hardly be useful for your career.
автор: Gabriel P•
Useful videos and assignments, although some are very unclear. The last assignment has really bad documentation of what everything is actually doing and it includes setting up a ubuntu instance on AWS, chat bot on telegram and tons of other stuff, and there are very vague explanations of how everything is linked together. But in the end when everything was done, i felt like ive learned a lot, including things out of the scope of teh course.
автор: Milos V•
Definitely best course in the Specialization! Lecturers, projects and forum - everything is super organized. Only StarSpace was pain in the ass, but I managed :)
автор: mirgahney h a m•
Amazing course lot of stuff to learn and it is very interesting with a good mix between the classical tools of NLP and more advanced one like deep learning.
автор: Weiyi C•
Only the initial contents were apropriatly explained. They only show mathematical models with out giving good examples in depth to explain the models, just mentioning that they exists, not a very practical course.
автор: Ankit B•
The content covers a lot of aspects of NLP but fails to deliver it in a comprehensive and detailed manner. Lectures are not that engaging.
This course is good if you want to get familiar with different methods/tasks in NLP and then supplement that with external readings.
автор: Mohamed A H A M•
Unlike common beginner online courses, this course delves a bit more into the math and explores state-of-the-art papers at the time of publishing the course.
The assessments are appropriately challenging unlike other courses which you can pass without doing any effort. In this course the quizzes encourage you to get a piece of paper and do a small calculation. I believe that is what online courses should be like.
I think the programming assignments should be upgraded from TensorFlow 1 to TensorFlow 2 though, or even PyTorch.
автор: Abhishek S•
One of the best courses I took from coursera. Good mathematical knowledge, resources provided are related to current research. Assignments are more than expected.
автор: Glorian Y•
Very good course with tough assignments. Expect to take more time to digest the material on the video and work on the assignments
I would not recommend this course to anyone who wishes to learn Natural Language Processing. The instructors do try to explain well, but the content itself is not structured well at all.
While mathematical knowledge is necessary, this seems to be the primary focus of this course, as opposed to practical experience. The assignments are ridiculous, with no relation to the course material, and concepts that require learners to spend hours trying to decipher. If documentation and wiki links were all that were required, people wouldn't be paying to learn the subject.
All they asked was to execute normal Python mathematical functions, with no explanation as to why. I do not feel good about completing this course. It felt like a complete waste of time and I have not gained any experience at all.
автор: Geovana R S S•
Well, if you already know NLP go ahead. But if you're trying to learn it, DON'T take this course. Because (a) lectures only talk about papers superficially and algorithms are not explained didactically; (b) lectures are so difficult to understand that you probably will need to see it many times to answer tests; (c) practical assignments with bugs (d) practically no written content, only videos.
автор: Robert H•
Explanations are not in depth. Problem sets introduce knowledge not provided in the lectures. This course needs to do in-depth explanations of concepts before throwing them onto a problem set.