great

Good!

great

Nice!

good

nice

good

good

good

Good

good

good

good

Wow

ok

This was, in general, a good course. The instructor was very clear in what he presented, and gave a good overview of Social Network Analysis. However, there were several issues with the AutoGrader that did not get fixed until late in the course and the PowerPoint slides for the lectures were also very late in getting posted (they were not available for most of the programming assignments). So, I think this course was launched a little early. Still, these are problems that you might expect to see the first time a course is taught and should not affect future students.

The bigger complaint I have on the course was that it was a very gentle introduction of the topic with only a quick overview of the subject. The lectures themselves concentrated more on a litany of various measures and metrics to characterize networks and could have benefited from a broader examination of real networks in the real world. One of the most interesting topics was a very quick overview of plotting for network diagrams, but this was never followed up with a programming assignment or other aspects to give us practice using the techniques described. This course would benefit from 2-4 additional weeks of material and more programming assignments, IMO. The network graphing lecture, for example, could have been reinforced with a peer-graded assignment to have us produce 3 or 4 types of graphs of various networks.

Overall, though, I was pleased with this course and the entire specialization. I would definitely recommend it to others.

This was a good course. I learned a good amount about network analysis and the python library networkx. I can envision using what I learned in my job. However, of the five courses in the Applied Data Science with Python Specialization I felt this was the weakest offering.

1. The Title. While the majority of the examples and exercises were focused on social networks, there's little in the course that is really specific to social networks. The course applies to any kind of network that can be loaded into networkx.

2. Trim the Process Descriptions. Too often the lecturer would say things like "Node A has degree of 3 because it is connected to three other nodes. Node B has a degree of 5 because it is connected to five other nodes. Node C has a degree of 4 because it is connected to four other nodes." For such a simple concept, that many examples aren't needed.

3. Provide On-Screen Example Files (my biggest gripe). In all of the previous courses, when the lecturer gave code examples on screen, there was a corresponding Jupyter notebook with those examples so the learner could follow along, and keep the notebook as a handy refresher of how to interact with the library. None of that was provided in this course.

Extremely good introduction to network analysis. The course heavily relies on NetworkX, and doesn't require extensive programming knowledge - with the help of Google, you may easily solve all problems. The lectures were well structured and easy to follow. Having said this, I have found 2 major drawbacks: 1. I would really appreciate some external references so that I could get a theoretical introduction to the materials taught. 2. The last assignment required machine learning, which was not taught in this course. With the help of the forums and a bit of googling, it is easy to get full mark, but perhaps the authors could include such background in the provided notebooks?

The explanations were very really good and clear but not enough to complete the assignments. The assignments were over the top in difficulty. The hardest in the entire course program. That is the only reason I took one star. It was because I felt that the classes did not prepare for the assignments. Or, assignments should have a more clear explanation of the steps to be taken in order to complete them. Definitely we should look for answers ourselves but not being able to clearly understand each step throughout the assignments really limited my research area and increased my frustration.

I started the course only because it was part of the Specialisation, but I am glad I did because the topic is actually very interesting! This course covers the basics. The lectures are very well structured, quizzes are suitably challenging, and the assignments are interesting while not terribly challenging. You'll apply some of the machine learning concepts from course 3 in the final week's assignments, which I though was a nice, round finish to the Specialisation.

Learnt considerable amount about social network from this course, as introductory level, materials (lectures and assignments) are well-prepared, much better than course 4 (text-mining). Assignments are not too hard, probably has relative good foundation from previous 4 courses. Auto-grader is a real pain in this specialization (course 3, 4 and 5), need to go through thorough test before release.

Do not consider this specialization as intermediate level.

Its a amazing course for beginners with little Python experience. The lectures and quiz are simple and assignments are really challenging. If you are looking for Social Networks course which covers nook and corners of Social Networks Analysis then this course is not for you.

Pro: Required interpretation of methods presented for application on assignments without explicit direction. Required application of knowledge gained in previous specialization courses.

Con: Explanations of social network analyses were limited in number and shallow in coverage.

The course starts off a bit slow but gets you used to the NetworkX module. The last exercise is a pretty neat culmination of the this course and specialization. It would have been cool for it to also involve text mining, but I enjoyed it and the course in general.

The course provides a great introduction to graph analytics, I consider that the social network applications are very sparse or missing in action altogether. Nonetheless, overall great content and practice of extracting information from networks with Python.