Perspectives on data science

DAPS - Data-Analytic Problem Structure

business understanding

Models for structuring big-data and data-analytics projects typically start with a definition of the project’s goals and the business value they are expected to create. The…

Jeroen de Mast, Joran Lokkerbol

The Design Philosophy of Great Tables

We’ve spent a lot of time thinking about tables. Tables—like plots—are crucial as a last step toward presenting information. There is surprising sophistication and nuance in…

Rich Iannone and Michael Chow

Portable dataflows with Ibis and Hamilton

composable data stack

data engineering

This post showcases how Ibis and Hamilton enable dataflows that span execution over SQL and Python. Ibis is a portable dataframe library to write procedural data…

Modern, hybrid, open analytics

composable data stack

data engineering

As a Python data user, I’ve wanted a more modular, composable, and scalable ecosystem. I think it’s here. Wes McKinney released pandas c. 2009 to bring dataframes into…

Getting into Python

How to get into Python, the most widely used programming language for data science.

Data visualization with Vega-Altair

I am a big fan of the Vega-Altair ecosystem for data visualization, because it not only helps me in creating appealing, interactive visualizations, but it also helps me to…

Analytical problem solving

I co-authored a peer-reviewed paper that compares the different modeling approaches for analytical problem solving.

Daniel Kapitan, Jeroen de Mast, Stefan Steiner, Wim Nuijten

Stop aggregating away the signal in your data

exploratory data analysis

By aggregating our data in an effort to simplify it, we lose the signal and the context we need to make sense of what we’re seeing. Originally published on <a…

RIP correlation. Introducing the Predictive Power Score.

exploratory data analysis

We define the Predictive Power Score (PPS), an alternative to the correlation that finds more patterns in your data. Originally published on <a…

Florian Wetschoreck

The Precision-Recall Plot Is More Informative than the ROC Plot

performance metrics

An introduction to performance metrics for binary classification. Originally published on <a…

Paul van der Laken, Daniel Kapitan