Welcome back to the round-up, an overview of the most interesting data science, statistics, and analytics articles of the past week. This week, we have 4 fascinating articles ranging in topics from data presentation to ensemble learning. In this week's round-up:
- How P&G Presents Data to Decision-Makers
- The Three Cardinal Virtues of Ensemble Learning
- Why? Why? Why! A Lesson for Data Science Teams
- Tableau Files For IPO
This Harvard Business Review article gives an interesting inside look at some of the ways Proctor and Gamble presents information internally about the performance of its business units. Emphasis is placed on presenting data in ways that help establish a visualization language within the company so that decision-makers can instantly understand what it is they are looking at and what it means, as this speeds up the time it takes to make strategic and tactical decisions. The article goes on to describe in more detail how the company does this and provides some examples of the types of visualizations they use and what tools they used to implement these processes.
I'm a big fan of learning resources that take complex topics, break them down, and explain them in ways that ordinary people can understand. And that's what the folks over at BigML have done with this post about ensemble learning. In the post, they try to explain why ensemble learning is beneficial versus running single machine learning models. The author breaks the explanation down into three virtues - Temperance, Prudence, and Fortitude - and uses each of these to explain different aspects of ensemble methods. For those looking for more in-depth literature, the post also contains a link to a more academic paper on this particular subject.
This is an interesting O'Reilly article advocating data science teams asking the right questions in order to discover what's really needed when working on a business problem. The article provides some guidance on how you can do this. The authors also cite a couple examples from their own experiences where what the company they were working for really needed was significantly different than what they had initially asked for, and they describe how it took a lot of necessary question-asking to get to the root of what was required.
Earlier this week, Tableau - the data visualization and discovery software company - filed for an initial public offering (IPO). Tableau has been growing rapidly over the last few years, and this announcement confirms their intent to continue that growth with the influx of capital the IPO will give them. This Forbes article provides some additional details about the IPO and some background on the company.
That's it for this week. Make sure to come back next week when we’ll have some more interesting articles! If there's something we missed, feel free to let us know in the comments below.