Data is the new science – We just got certified for it!

Ten weeks of Just-Data Science journey! Read in [est_time]

Twelve weeks ago, six Just-Consultants stepped into the world of data science. Hoping to get a better understanding of what it’s all about and figuring out if it is more than just a hype, this group started a Data Science course at DIKW in Nieuwegein. As you might have read in our previous posts from Viveca Cohen and Eline Bangert, the training covered a lot of different areas within the data science space. The first weeks were all about the basics: introduction data science, R, Python and matrix algebra. After that the main algorithms and methods in data science were covered: regression, clustering, decision trees, text-mining and lots more.

The Exam
After studying the course material, homework exercises and a data science book, the students were tested for their knowledge. The test consisted of the following parts:

  • Theory (50 questions)
  • R assignment
  • Python assignment

Data Science teams
Next to that, groups of max. 3 students, had two weeks for working out a real data science problem, provided by the students themselves. In a presentation the students were requested to explain this problem and the approach taken to solve it.

The Just-Consultants were split into two groups. The first group tried to determine the best way of investing money into the electricity network in order to reduce the number of / duration of disruptions (a relevant case for a large electricity provider in the Netherlands). In order to answer this question, historical data of all disruptions that took place since 2004 was analyzed. The second group focused on a dataset that contained all queries (views) on the Just-BI HANA system. By using clustering techniques, they grouped queries that are most similar: on the same table holding mostly similar columns. Using this information, it is possible to reduce the number of views which will reduce the maintenance costs.

The road to success is not easy, but with hard work, drive and passion, it was possible to achieve the Just-Science dream!

After 10 weeks of hard work, homework and a lot of learning, all six of the Just-Consultants managed to pass the exam and the assignments. Meaning that right now Just-BI has their first consultants that may call themselves ‘Certified Data Science Professionals’. Congratulations to the good work Eline, Ralf, Scott, Shervin, Viveca and myself!

 

This article belongs to
Tags
  • algorithms
  • CDSP
  • clustering
  • Data Science
  • decision trees
  • DKIW
  • Python
  • R
  • regression
  • text-mining
Author
  • Ivo Moor