The Data Skeptic Podcast features conversations with topics related to data science, statistics, machine learning, artificial intelligence and the like, all from the perspective of applying critical thinking and the scientific method to evaluate the veracity of claims and efficacy of approaches.
This first episode is a short discussion about what this podcast is all about.
(upbeat music) - Welcome to the inaugural episode of the Data Skeptic Podcast, a podcast which discusses the applications, techniques, breakthroughs, news items, and claims that fall under the umbrella of data science. This episode will have a much shorter and different format from typical episodes. I wanted to take a moment to introduce the show in such a way that could serve as a reference point for people who asked the question, you know, what is this podcast or why am I starting it? The particular focus of the Data Skeptic Podcast will be about exploring topics that fall under that vague heading of data science. In particular, I want to explore the plausibility of claims, celebrating the breakthroughs and the good science. Well, occasionally chastising the sensationalism and the "magic proprietary black box" and "phenomenon." I've started to notice a certain willingness for people to believe the unbelievable when there's a hand-waving answer that it says it's all based on data and algorithms. My motivation to start the podcast is not only to frankly have an excuse to have good conversations with some of the best and brightest in the field, but also to hopefully present topics in a way that makes them accessible to listeners without a computer science or a statistics degree, and to encourage critical thinking of any claim one stumbles upon, whether they be data-centric or not. If I were to describe myself by a couple of adjectives, certainly one I would include would be skeptic. Cloquially, this word is often used sort of synonymously with cynic or a contrarian, but I think this is truly a misuse. When I say skeptic, I mean it in the same way people like Carl Sagan, Penn Gillette, James Randi, Eugenie Scott, Steven Avella, and Brian Dunning mean it. Skepticism is all about applying critical thinking and having an appropriate level of belief or doubt of a claim that's commensurate with the available evidence that support that claim. In future episodes, you'll certainly hear me throw around the term "baisy" in a lot. And although I won't define that here, I'll probably end up defining that in some of what I call my mini episodes that I'm going to include in the feed. So I'm aware that not every listener is going to have a technical background, and at least I hope not, I want this to be very accessible podcast, yet I know I can't stop every conversation to define what type two error means or what normally distributed means. So I'm going to occasionally do these short episodes mixed into the feed that I'll call mini episodes that cover topics like that, and Bayes' theorem will certainly be an important one, because that's a very useful tool for how to assess evidence. Anyway, my plan with the Data Skeptic podcast is to primarily do long-form interview-ish discussion programs that explore news items and particular research about topics that's under this vague collection we nebulously refer to as data science. So if you're someone well-versed or just casually interested in topics like algorithms, search engines, information retrieval, statistics, data mining, forecasting, regression, image recognition, natural language processing, artificial intelligence, decision theory, game theory, and so on and so forth, I hope you'll join me and all my guests for an exploration of these topics. I'll do my best to make these concepts accessible to everyone regardless of their background, and to get all my listeners to realize that there truly is an unreasonable effect in this of data, but that doesn't mean that anyone with a fancy algorithm will certainly hold a solution to whatever problem you're facing. I intend to keep show notes, sometimes data files, plots, or short amounts of code that relate to each episode, and make them available on a public GitHub page. If you don't know what GitHub is, don't worry. I'll do a mini-episode about that coming soon. So why not just head over to datasciptic.com, which will be the greatest starting point for anything related to the show? I'm Kyle Polich, thank you for listening. (gentle music) (gentle music) (gentle music) (gentle music)
