Nice

good

ok

ok

e

V

In general, an excellent course, taught by competent professors. I believe that in the main this course does very well in achieving its objective of knowledge transfer. However, having experienced it, there were parts where the professor was demonstrating a topic using a video presentation showing him operate a process or screen sequence on his computer. These aspects, like virtually all the material on this course, are of a technical nature and contain many important details. As such, to help complete and re-enforce their learning, students require something like a sequence of slides that they can print out and retain for revision and future reference. In certain parts, the provision of the printable screenshots in the form of slides was absent.

An important theme of the course and Data Science in general is "Reproducible Research". What I'm arguing for here is, "Reproducible Learning Materials" covering a complete course, not only parts of it. Admittedly, it was only a very small proportion of the course that suffers from this defect. But I would not like it to become the norm in the future. As a suggestion, it could be possible to author a lecture using HTML so as to combine the verbatum lecture text with every slide/screenshot image embedded in its right position within the lecture.

I notice Coursera courses have also moved away from the weekly lists of individual lectures together with their links to .txt, .mp4, etc. files. The new presentation keeps you submerged within the flow within each week's series of lectures. One has to 'click out' in order to watch your progress and then re-enter the lectures at a resumption point. I prefer the previous navigation structure in order to access lectures and materials. Printed learning materials are also important for me, in addition to the video lectures. The latter are of course vital as the medium for the initial exposure of the material.

I think this one is the best from the Specialization as it is the most Practical one (more than Practical Machine Learning).

Can be taken without the others if you have basic experience in R and want to learn about cool R applications.

The reason I don't give full rating is for not having practical assignment every week. So there wasn't enough effort put into the course. Of course, we can do optional homework and make more applications, but assignments like these should be mandatory. There is no package building and no swirl course building - So why do we have week 3 and 4 at all? The quizzes are also laughable - no knowledge testing at all.

So although I liked this course from the Specialization the most, I still can't give full rating because of the mentioned issues.

The skills taught in this course are fantastic and I'm sure using them will blow my colleagues' minds away. However, I must say that the lectures on Rcharts and other interactive plot builders sound kinda sloppy, poorly prepared. I know the documentation for those packages is bad and it takes effort to figure out what they do, but that is precisely why a well-prepared lecture would be so useful. I would also talk about license, since we have been dealing with packages that are completely open for use, but these have some restrictions.

Taken this course in its old fashion style. Now reviewing the new design was a little bit displacing, but I ascribe this to the fact I've done all the specialization courses in the old design.

However the structure of the course is quite good. Some typos were reported, as well as a bug on the unanswered questions in quiz 3. Main worst point was the missing format of several text boxes. I would have appreciated paragraphs, bold and italic, some links, picture, not only raw plain text.

Overall review is nonetheless over the average.

Great course, I just missed some material on distributing data products as files or objects. Data Science environments are getting connected to traditional BI-environments more and more, now that organisations are getting more used to DS. So it is starting to be important to also deliver data products as files to the e.g. data warehouses, ArcGIS, or open data platforms. I know this is mentioned in Getting and Cleaning Data. But some further elaboration would be nice.

This is an excellent course that will teach you plethora of new things! The only gotcha is that things move too fast in the world of Data Science. Some of the topics and code might not work exactly as shown when you take the course, and many things will change. It's up to the student to make-up for the missing pieces, but I guess that is the only caveat when you work in a still-evolving field.

This course gives a good introduction on how to develop an application.

It gives all the available tools in the field out for us to try and use them.

The course is enjoyable and not stressful. I find the assignment as meant to get us to do the project and not really there to fail us. It is difficult to fail to course as only the minimum is required.

Analysis is useless if done for its own sake. Once you have found something interesting the challenge is finding engaging ways to share your insights. This course is a bit scattered since it covers so many different ways of publishing and presenting data, but it is a really nice survey of what is out there.

I lost myself a little bit, because the materials were a little over the place, we did shiny in the first week lectures and only in the last week the project was about it.

Same with leaflet, I think ti could be structured differently.

But i find them very relevant so that is the reason why I rated it with 4.

You will need to be familiar with the subject in order to keep up with the assignments or put in a large amount of time to learn. The forums are helpful and other students are great. Visit the shinyapps.io website for Shiny Apps and slidify research. Rigorous and rewarding when you put the work in.

It is a nice class but many of the topic were already covered on previous class of the specialization. If this is the only class you will take, it is okay but it you are taking the whole specialization this is like doing few (not all) things again

Good course, but it was a little light on content. Probably the easiest month of the specialization aside from the first one. The assignments were too easy too pass. Good intro to some very interesting packages (leaflet, shiny, google vis) though.

I enjoyed the class, but was frustrated when it came time to get my Slidify to work on GitHub (since RPubs wasn't publishing them correctly.) I had to convert it to RPres (which was easy, thankfully) in order to get my project submitted in time.

I liked the course and found it informative, but wish there were more stuff on Shiny Widgets and Input/Output/Render topic. R Shiny tutorial is pretty good, but I was hoping more relevant info about those topics from this course.

This was just a quick overview of different technologies out there to help creating various types of interactive graphics in R. But I would have preferred if at least one of those technologies were explored in more detail.

This class includes a lot of introductory content an a good measure of hands-on practice. It also provides a great amount of resources and reference material for the learners to expand their knowledge further.

Nice course and good classmates. It's very focus oriented and gives very good idea of Shiny, Rmarkdown, plotly and how to publish documents on github, Rpubs and other online sites. I learnt power of R

Compared to the other classes in the JHU Data Science specialization, this one is pretty laid back. It's useful information, and teaches a few nice tricks on how to present data analysis results.

This course lacked required information of help to get started. Now thanks to some posts by mentors i was able to successfully complete the Capstone project. Overall a very good experience!