General Topics in Data Science

History, Approaching Data Science Problems, and Other General Topics

History of Data Science

While many of the elements of data science have been around for a long time (even the idea of machine learning, AI, and the perceptron date back to the 1950s), the profession of data science / data scientist seems to have emerged around 2008.

This first article below goes through the history of data science, including its consituent parts - statistics, data analysis, machine learning, data science tools.

As a disclaimer, while most of the posts on this website are not connected to the curator of this site, this is one post authored by myself.

With two big components of modern data science being statistics and machine learning, it is useful to think about the differences between the two, especially to understand different peoples’ approach to a problem. The way a statistician-by-training might approach a problem will likely be different than the way a machine learning practitioner will approach that same problem.

Approaching Data Science Problems

While it may seem strange to discuss this when data and science are in the name, there is some amount of intuition that comes into play in data science, and certainly when analyses and results are handed over to people / teams who are more on the business side of things.

Here below, I summarize three blog posts that try to tackle this idea of intuition in data science. It is important to think about where intuition most often is relied upon and how much practitioners rely on it.

Rules-Based Engines to Machine Learning

Two articles below describe considerations when considering whether to transition completely from a more traditional rules-based approach to machine learning, or a combination of the two.

You can read about a recent, real-life use case of combining rules with ML from Netflix in the section on neural networks.