Introduction to Data Science

Introduction to Data Science is a MOOC offered by the University of Washington on the Coursera platform. Introduction to data science is a misleading title for this course because it is not introductory level and it does not have a sensible flow that builds from one week to the next as you would expect from an intro course. Instead, the course acts as more of a data science sampler that introduces new topics each week that often have little to do with material covered in previous weeks. Lecture topics include relational databases, relational algebra, SQL, MapReduce, No SQL, miscellaneous topics in statistics, machine learning, visualization and graph analytics. If that sounds like a disjointed smorgasbord of topics, it is. To make matters even more complicated, the programming assignments use three different languages: Python, R and SQL. This course is best suited for those who have some exposure to Python, R, SQL and statistics.

If you have the appropriate background knowledge, this course touches on many interesting topics and while the lecturer's delivery is not great, he is quite knowledgeable and the material usually isn't too hard to grasp. Although the homework assignments require different languages and may take you a while to complete, they are rewarding. For instance, you'll work with real Twitter data you capture from the net, implement MapReduce operations in Python and participate in a machine learning competition on Kaggle.com.

Introduction to data science is likely to be frustrating to those expecting a general intro to data science. The course jumps around too much and uses too many different tools to be a good first course in data science, but the breadth of topics covered and programming assignments make this course worth a look if you already have some exposure to data science or the tools the course uses. If nothing else, you can skip through the lectures and watch sections that are of particular interest to you.

I give this course 3 out of 5 stars: Satisfactory.

The lectures by Prof. Howe were fun and of good length to watch on a busy schedule. They supported the projects quite well and I think that the professor did an excellent job with this course. Unfortunately, some topics like Machine Learning lacked a programming assignment to support and apply the theory, so we took them as a theoretical introduction. The Visualization lectures were of particular interest for me since I am an intermediate level Tableau user. They came from the basic concepts to some advanced tips, really useful for users of all levels. Finally, a presentation of two products that goes beyond the MapReduce paradigm: Wibidata, with a good product presentation, and Datameer lecture was more of a sales pitch than lecture, which some students considered not appropriated (including me).

I am very positive about the course. I had background in MapReduce, graph visualization and programming, but the level of the course covered from basic to more advanced techniques in all the areas. The assignments reinforced the lectures and were appropriate. On the other side, some issues with the autograding system, complains from some students of the peer evaluation grades, and some delays that the staff and Prof Howe replied in the forums to help in some situations gave some negative impression. According to the forums, most of the students are satisfied with the course and some of them are expecting a new session.

This course gave an excellent platform to channelize my preparations for the industry. The very first week we had to do a project in Python, where we accessed Twitter API and did sentiment analysis on a sample of live tweets. Now people working in this area know that sentiment is an extremely important application in big-data world. Within a week, we're working on the topics of current interest to the industry. The lectures by Prof. Howe were very well thought and delivered with supreme clarity. The topics selected for this 8 week course were very relevant, we were taught basics of MapReduce, Relational Databases, NoSQL, machine learning, data visualization, graphs etc. We did exercises in MapReduce implemented important algorithms in Python.

The course was very modern in its treatment. We were required to participate in a Kaggle competition and there was an optional big-data project which required setting AWS and programming in Pig. I also worked on a real world project, Prof. Howe invited organizations all over the globe to post interesting projects for the students taking the course. Completing such a project was optional, though one could submit the project for peer review. Moreover there were guest lectures by experts from industry and academia.

Professor Howe's lecture style is not always engaging, and a lot of material is covered. There were often over 3 hours worth of lecture material to review during the week. Along with following links and reading supporting papers, this left very little time for the assignments themselves. Prof. Howe did a good job of introducing and comparing a range of current technology choices (particularly the comparison of different database technologies). As a data science newbie, I would have liked a bit more information and emphasis on use cases for different types of databases.

The level of the first assignments was not commensurate with expectations from the course description. I learned a lot, more than I expected, in fact, and I can now implement a matrix multiplication in Python, SQL, or MapReduce based on the homework assignments. The auto-grader for the 1st assignment never did accept my answer for the final part. It also didn't give sufficient feedback for me to solve the problem, which probably had something to do with text encoding, but was very frustrating none-the-less. This sort of issue doesn't teach much. Save your perseverance for things that matter.

While course assignments are invaluable and aimed to build programming/analytical skills critical for understanding big data processing, machine learning, and data analysis,

the video lectures are sketchy, boring, and not sufficient for assignments.

Staff wasn't engaged or present on forums and there were no even TAs.

Prior programming experience, especially dictionaries and machine learning algorithms, stackoverflow and google, helped me to complete the course.

My favourite part of the course was participation in Kaggle competition. I felt okay doing it just because of the knowledge gained by taking Jeff Leek's and Andrew Ng's courses previously.

Other assignments included twitter sentiment analysis in python, in-database text analysis including keywords search, MapReduce algorithms, Poker Analysis with JSMapReduce, optional large scale data processing using AWS, data visualization with Tableau, and optional real world project that connected students with organizations that is great and innovative for MOOCs.

It was a mammouth ambition to cover the basics of data science in fewer days than Phileas Fogg took to travel around the world. The lectures, at time, felt disjointed, since there was so much material to cover. The breadth of coverage was phenomenal, though, and I am now at least comfortable with the cohort of big data jargons like Hadoop, HBase, CouchDB, Pregle, and MapReduce.

The first few weeks were focused on programming assignments using Python and SQL, both of which I had some familiarity with. The exercises were challenging enough for me to learn new tricks of the trade but didn’t feel like I was drowning. As the course moved on to MapReduce concepts, I started to struggle. But again, it was just the right level of challenge for a newbie that I felt compelled to motor through the quizzes.

DO NOT TAKE THIS COURSE.

IT'S A COMPLETE MESS!!! SO MANY TYPOS EVERYWHERE!!! THE TEACHER JUST THREW A BUNCH OF IDEAS TOGETHER AND CREATED PRESENTATIONS A FOUR YEAR OLD COULD MAKE.

BILL HOWE DOES NOT KNOW HOW TO COMMUNICATE IN GENERAL AND DID NOT PUT ANY EFFORT INTO MAKING THIS COURSE. DO NOT TAKE THIS COURSE IT'S A TOTAL SCAM! I JUST LOST $15 ON AWS TRYING TO DO HIS POORLY LAID OUT HOMEWORK ASSIGNMENTS (THAT'S WHY THIS REVIEW IS ALL IN CAPS!!!). I WAS 90% COMPLETE WITH THE ENTIRE COURSE SERIES (I WAS BEING STUBBORN) UNTIL I FINALLY HAD ENOUGH OF THIS MEDIOCRITY.

A MOOC SHOULD NOT BE THIS UNENJOYABLE!!!

CARLOS GUESTRIN AND EMILY FOX'S MACHINE LEARNING COURSE IS A MILLION TIMES BETTER AND THEY COME FROM THE SAME UNIVERSITY???

P.S. IF BILL HOWE EVER READS THIS REVIEW I HAVE A MESSAGE FOR HIM.

THINK YOUR LIFE OVER...TEACHING ISN'T FOR EVERYONE!!!