So we’re into the fourth instalment of the John Hopkins Data Science Specialization on Coursera.
Perhaps I am being harsh, but I didn’t enjoy this course. I just found it a bit dull. The first two weeks are basically just drawing charts or plots; or urrgh “graphs”. I prefer to think of this as a graph (:Node)-[:Rel]->(:Node), but hey I’m a Neo4J fan-boy.
The final week comprises of two lengthy case studies videos. Week three is the “look what you could have won” moment when Roger just touches on some interesting topics like hierarchical and K-Means clustering. Unfortunately these topics are not part of either the assessed quizzes or course projects.
There are no Swirl exercises either, if you’ve read my reviews of the other modules you’ll know I have the optional extra credit exercises down as a real plus.
There are two assignments for this module; I’d say both of them can be done in the time it takes complete one from the Getting and Cleaning Data module. Not particularly challenging, mainly limited to plot so-and-so data by X and Y.
The two quizzes in the module are fairly standard, answer ten multiple choice questions with two re-attempts available. Again not as taxing as other modules as they require you to do very little with R Studio or the Console.
Pretty short review compared to my others, but I don’t think this module deserves more.
Next up is Reproducible Research, I just hope we’re not reproducing what we researched in this class or I might lose will to complete the class!