Very good course from deeplearning.ai team , you will need some background in ML and Python. The support on the coursera forums could be better.

Academically, a step down from deeplearning.ai's previous courses. In terms of technical quality, the media could use improvement, particularly in normalizing audio levels and ensuring a perfect acoustic setup for all lecturers. Coursera should hire some acoustics and motion pictures experts and task them with improving "production values" (look it up).

Shortcomings notwithstanding, this is still a great class and a "must take" for any aspiring NLP expert.

HW check is just awful. I need to use exact these functions (i.e. squeeze instead of ravel).

Moreover, one could complete homeworks without referring to the lectures at all -- zero challenges met.

The last HW is not balanced (11 tasks -- too much compared to the rest).

I would suggest the course makers to redesign the homeworks dramatically.

Thank you first of all. I think the length of videos could be longer for better understanding of concepts. Specifically, I think in Week 4 there are a lot of concepts like ( Locality Sensitive Hashing) that could be better explained.

The assignments lacked clarity. The huge number of variables used for seemingly trivial intermediatory task is really confusing. Also, the assignment documentation and instructions need to be enhanced,

The exercises and assignments seemed to place more emphasis on the coding and less on the theory and study of natural language processing per se, imho.

T

h

e

c

o

n

t

e

n

t

s

a

r

e

a shallow, the lessons doesn't go into details. The teacher only shows what's needed for understanding the week exercises. I won't dare say it's even basic level.

Really too elementary. You can skip almost the whole course if you attended the DeepLearning specialisation.

By far the worst course I have ever done.

The 'lectures' are just a few minutes long and as expected are very thin. Moreover, the majority (yes > 50%) of the lectures are spent explaining basic linear algebra (much of which is high school level) and hardly any time is spent on NLP itself. Most of the core NLP concepts are barely mentioned and it is obvious that the labs were written by someone else than the person who wrote the lectures as terminology gaps / differences exist and as the content is very thin it's impossible to put two and two together.

The readings are simply annotation summaries of the already brief lectures. They add no value.

The labs do not support learning in anyway. They are simply a 'fill in the gaps' exercise where > 50% of this would simply be figuring out which variable you need to index inside a loop. It is a joke.

The only good aspect of this course is the slack community and is the only thing I can recommend. If you can sign up for the trial for the course / specialization and participate in the slack community. Apart from that, the remainder of this course is a huge waste of time: you won't learn NLP here, you will simply have a guided python tour with NLP problems as content.

I came to this course motivated and excited. However, after having taken the machine learning specialization taught by Professor Andrew Ng, I was disappointed by the quality of this course. Not only does this course go too fast and doesn't explain the intuition and reason behind concepts, the labs are outdated and have terrible unit tests. I work by day and study by night, only to find a programming assignment (week 2) that has terrible unit tests. It tells you that 12 tests passed and 3 didn't. Why? Who the hell knows why? Which defeats the concept of a unit test. Then I found myself debugging the unit tests themselves to understand why the they fail. I mean, calculating the loglikelihood is straightforward, and yet, it said that the tests failed. Go figure! I do NOT recommend this course!

I am waiting for my certificate for over a month now, there seems to be the problem with the grader that wont mark my assignment as passed. I reached for support multiple times now and no matter how easy the problem is or how important to me it is, nothing is done. Pretty dissapointed. In general the course is great.

Very disappointing.

The course aims at teaching some copy/paste recipes for algorithm rather than an understanding of the theoretical basis behind these algorithms. Plus, the assignments are about dealing with numpy/python syntax, there is no thinking about how to implement an algorithm as everything is pre-entered.

there is no note provided

the programming questions and instructions are so ambiguous

this led to much confusion

I enjoyed the course very much! I would say, there is still some things to improve, like the treatment of the implementation of the hash-local search, or some typos here and there along the way in the notebooks. Another suggestion is to postpone the implementation of SGD until after you get some warm-up in numpy during week 2-3. All in all I quite enjoyed the approach to be able to build the tools myself with numpy. This way, it demystifies the use of pre-built packages and lets the learner indeed understand what is going on under the hood. Of course, there is always a trade-off between being precise and being easy to understand. I think nevertheless that this course is spon-on in this trade-off optimization task. ;)

This is an amazing course for beginners in the field of natural language processing. It starts with the very basics of machine learning and natural language processing. Exploring sentiment analysis using two different approaches one is a frequency-based approach (involving logistic regression) and the other is a probabilistic approach (involving naive Bayes approach). This course also gives a glance of vector spaces and techniques to convert multidimensional data to lesser dimension data using approaches like PCA. I learnt new topics like locality sensitive hashing during the last week of the course. It was a fun and engaging course overall :)

I think this course really opens up the student to the field of NLP, the concepts shared are simple enough to understand, and the codes shared in the lectures / tutorial labs are really useful to understand how the theory is translated into python codes and applied into working models. I love learning NLP! When I am learning NLP I am Happy indeed with DeepLearning.AI, thank you Professor Andrew Ng and et. al. The Training Instructors do explain the concepts and application codes very well. My heartfelt thanks and gratitude to this course!

I really like the way the course is structured and the way it is taught. The language is clear and it goes at a good pace. I have completed the Deep Learning Specialization as well and felt that this course had a better segregation of material/videos. The short videos make sure you don't lose focus or get bored. Content-wise, the course is great for understanding the initial steps required for approaching NLP problems. Highly recommend it and I look forward to completing rest of the courses in this specialization.

No matter who you are or who you have been, you can be whoever you want thanks to this course you will understand in great detail the Natural Language Processing with Classification and Vector Spaces. The teacher rigorously explains its content by putting you in a real situation and the key concepts are very well explained in clear the academic rigor is there now I can affirm without reserve that I understand very well several concepts which were strange for me thank you to coursera and deeplearning.ai

This course had helped me become familiar in natural language processing. Before taking the course, I feel that NLP is already like a plug-and-play thing due to deep learning. However, it had helped me understand the importance of preprocessing especially to get the right embedding. Additionally, it also help me understand a glimpse of doing sentiment analysis and Machine translation. Indeed, I still have many things to learn but this course is a great way to be introduced on the NLP.

Great intro to the world of NLP. For those with a background in python, linear algebra and statistics (my case) course will take significantly less time, as you can skip some of the videos and rush through easier notebooks. Assignments will make sure you haven't overlooked parts you don't yet understand. In my opinion comments and hints sometimes lead by the hand a little bit too much, but maybe I'll change my mind when faced with assignments from next courses of the specialisation.

I had attempted other NLP coursse before attempting this course. I found this course is a good to people who are new to NLP or even new to machine language. The hand-on examples and assignments are the plus points of this course. Though the coding the assignment may look like "fill in the blanks" sort of exercise, I appreciate it since it helped me to understand the concepts. Planing to take up the next course in this series. Great instructors and program is well structured.

Amazing course overall, I have a few years of experience working as Data Scientist and still could get many valueble learnings from it. So whether you are a beginner or someone with some experience I would recommend it. The material is great, the Notebooks used for coding are very well structured, the lessons are very focused - I recommending doing some searches on the side to make sure you grasp all the concepts properly. Also the Slack community is very supportive.

I had an project about NLP and I was trying to find a course on internet. To be honest I tried another website's course but it didn't help so much. When I see that the videos are really short in this course I started to think like "What, is this joke? Will I really learn anything?". However I understand that courses teach so many things. I'm really glad to get this course and I believe that it has given me a lot for my future career.

I've done online courses before, including other deep learning courses. But I can't talk enough about how well organized and just far superior the deeplearning.ai courses are. The video lessons are short and to the point, the exercise sessions are challenging but not impossible, and the community of students and teachers are all very active on the message boards to help you get through when you are stuck. Can't wait to do the rest!

Andrew Ng and his colleagues have delivered another excellent course - these really set the standard for online courses on machine learning. The concepts and coding are explained very clearly by the tutors, and the exercises offer a suitable level of challenge. The final assignment involves building an automated translation capability based on word vectorisation and hash tables, ie a potentially useful/usable capability.