April MOOCs - Focus On Data Science
Written by Sue Gee   
Friday, 03 April 2015

Each new month brings new MOOCs and re-starts of existing ones presenting an ever widening choice. With one exception our selection of Computer Science classes for April are on data science.

Let's start with the odd man out.

From Nand to Tetris Part 1 is on Coursera, starting on April 11th for 7 weeks with 5-10 hours of study per week. It comes from the Hebrew University of Jerusalem and in it students will use a hardware description language (HDL) and a hardware simulator to build a modern computer system, starting from first principles. The course consists of six weekly hands-on projects that go from constructing elementary logic gates all the way to building a fully functioning general purpose computer. In the process, students will learn how computers work, and how they are designed..

Given the boom in interest in Data Science, see What is a Data Scientist and How Do I Become One?, the other MOOCs in this round up are on this topic.

If you are new to data science one possible starting point is a self-paced Udacity course, Intro to Data Science, that you can access for free or enroll in at a cost of $199 per month after a 14-day trial. The prerequisites for the course, which has about 50 hours of content in total, includes programming experience, preferably in Python, and a background in introductory level statistics:



This course is part of Udacity's Data Analyst Nanodegree which is a rolling program which starts next on April 7th at a cost of $200 per month after a one-week free trial. To gain a nanodegree you don't just follow course materials, rather you build a portfolio of projects.

The Data Analyst nanodegree consists of six projects each linked to a Udacity course as follows:

Project Title Preparatory Class

Predict the New York City Subway

Intro to Data Science
Wrangle OpenStreetMaps Data  Data Wrangling with MongoDB
Explore, Summarize, and Discover Interesting Insights from Datasets    Data Analysis
with R

Identify Fraud from the Enron Email

Intro to Machine Learning

Tell Stories with Data Visualization 

Data Visualization

Design aad Analyze and A/B Test

coming soon


Any of the preparatory courses can be studied for free but all you get that route is the instructor videos and their associated videos and as it is self-paced this lacks the MOOC experience of a cohort of students progressing simultaneously.

Recently the NSA let it be known that completion of an online data science certificate program such as the one offered by John Hopkins University on Coursera could be counted as the equivalent of one year's job experience by job applicants which is welcome recognition of its value. 



To earn the Data Science Specialization you need to successfully complete the Signature Track for all of its nine 4-week courses (see plus a Capstone project - an assignment that demonstrates the knowledge acquired from the series of classes.

All of these courses currently begin at monthly intervals with the next starts being April 6 and May 4. The total duration of the program, including Capstone project is 40 weeks with an workload that varies from 4 to 9 hours per week and a total cost of $490.

Coursera also offers a Data Mining Specalization from the University of Illinois and the third of its five courses, Cluster Analysis in Data Mining starts on April 27,

The Coursera platform also has some standalone courses on specific aspects of data science. One that has just started on April 1st is Process Mining: Data science in Action taught, in English, by Wil van der Aalst of Eindhoven University of Technology. This is a 6 week course requiring 4-6 hours per week of study that  explains the key analysis techniques in process mining, which is described as:

 the missing link between model-based process analysis and data-oriented analysis techniques. 

 For a compact exploration of Big Data for those with a suitable level of prior education and experience MIT's Online X program, Tackling the Challenges of Big Data which sets out to:

survey state-of-the-art topics in Big Data, looking at data collection (smartphones, sensors, the Web), data storage and processing (scalable relational databases, Hadoop, Spark, etc.), extracting structured data from unstructured data, systems issues (exploiting multicore, security), analytics (machine learning, data compression, efficient algorithms), visualization, and a range of applications. 

Costing $545 it next runs from on May 5 to June 16 consists of five modules covering 18 topic areas in 20 hours of video plus case studies. Students who complete five assessments with an average of 80% will be awarded a certificate of completion.

Starting on April 20th Big Data: Measuring and Predicting Human Behaviour is a 9-week course with a workload of 4 hours per week. From the University of Warwick and on the Future Learn platform it is designed to be accessible to all and sets out to:

explore how the vast amounts of data generated today can help us understand and even predict how humans behave.

Participants are expected to gain an overview of the state of the art in big data research across a range of domains, including economics, crime and health and acquire some basic practical skills for data science by learning to write basic programs in R, create basic data visualisations and carry out simple analyses. 



Quantum Company Wins Digital Startup Contest

Quantum specialists Qilimanjaro is the winner of this year's Four Years From Now competition. The award was made at this year's Mobile World Congress (MWC) in Barcelona.

Stanchion Turns SQLite Into A Column Store

Stanchion is an open-source extension that adds columnar storage capabilities to SQLite, rendering it efficient for analytical reporting.

More News


raspberry pi books



or email your comment to: comments@i-programmer.info


Last Updated ( Monday, 22 June 2015 )