We have a special audio treat this week: a bonus episode of the Data [in the] Sandbox podcast miniseries! We explain what “artificial intelligence” and “machine learning” mean in a way kids can understand, using everyday examples like TV show recommendations, robot vacuums and math homework.
Those everyday examples have made “artificial intelligence” and “machine learning” familiar concepts not just to data experts, but also to the general public. But how familiar? And to whom?
Google Trends data can show us with a bit more precision how popular these terms are, how their use has changed over time, and even where they are more or less popular. And, of course, this is an analytic approach you could use with all kinds of keywords or phrases, for whatever topic interests you most. …
When I think of dangerous Christmas decorations, I always think of scenes like this one from “National Lampoon’s Christmas Vacation”:
And yet the most dangerous part of holiday decorating doesn’t involve electricity.
I downloaded and analyzed the U.S. Consumer Product Safety Commission’s latest 10 years of data on injuries involving Christmas-related products. Applying a bit of data science, we’ll find where danger might lurk among the sparkly lights and shiny ornaments. You might be surprised: One dangerous item is something you probably use year-round.
There are 3,917 injuries related to Christmas products in the dataset from the CPSC’s National Electronic Injury Surveillance System. Covering the years 2010 to 2019, the data include the injured person’s gender, age, involved body part, diagnosis, location (e.g., home, school), whether they were admitted to a hospital or not, and a brief text narrative of the incident. Although I retrieved data only for injuries including Christmas products, other involved products or key features of the injuries are also included (e.g., …
In this week’s Alter Everything podcast episode, guest Steve Mann from Alteryx partner Propel32 Analytics discusses the increasing importance of analytics in the mergers and acquisitions field in recent years. Data analysts and data scientists must constantly adapt to that kind of change, and there’s always something new to learn!
You may have heard of modeling techniques to predict the probability of churn for a customer, or to assess whether a customer will or won’t respond to an offer. But what about figuring out which customers might increase their purchasing — or could stop buying — as the result of a promotion?
Often we focus predictive analytics on modeling customer churn or a response to an offer (perhaps using logistic regression, as demonstrated in this excellent blog post). Uplift modeling takes a different tack. …
Have you ever abandoned a shopping cart in an online store and gotten a reminder email about it later? Your poor digital cart was stranded on a lonely server somewhere. But fear not, readers — we’re not abandoning you! Welcome to the second half of our introduction to market basket analysis.
In the first post, we covered some of the essential concepts behind market basket analysis, so check that out first if you’re not familiar with the basics. This post will show how to use this approach in Alteryx Designer. …
I cook green bean casserole just once a year. Although it’s kind of a culinary travesty, we still make it with Thanksgiving dinner for sentimental reasons. Its essential ingredients are green beans, canned cream of mushroom soup and — most important — so-called “french fried” onions (also from a can) sprinkled on top. All three ingredients often are grouped together in the grocery store around the holidays.
Whether you’re the kind of person who seeks out the spooky or not, guess what: You probably live near some creepy things.
To commemorate the season, we thought it would be fun to do some macabre mapping and petrifying prediction of spooky phenomena. Data science doesn’t have to be just for serious subjects! I’ll show you how I used Alteryx Designer, Python and the mapping package Folium to analyze and map these data.
To look at how spooky U.S. metro areas are, I created a (silly) Spooky Score for each area, based on the density of cemeteries and haunted places in each metro area, as well as the per capita UFO sightings and Bigfoot encounters. …
Ready to put your data science skills to work — to help others?
No matter what career level you’re at, you too can participate in “data for good” events and activities. Whether you’re established in or aspiring to a data career, there are plenty of opportunities for you to contribute. You’ll get experience in new domains, new portfolio projects, and new connections with other data enthusiasts, plus you’ll feel great about contributing to a good cause!
Caps, gowns, diplomas … and data!
Each student’s journey through a higher education institution creates lots of data. Recruitment, advising, retention, financial aid, administrative processes, assessment measures, course work, athletics and alumni activities all can be tracked in detail.
That data can be put to work in predictive models that advance institutional goals and aid student success. In addition to the effective use cases linked above, here are two more innovative ways researchers have used machine learning to make predictions in the world of higher ed. …
A research paper I read recently led me to consider: Could process automation not just empower humans by helping us avoid dull tasks, but also by fundamentally changing the way we think? Considering automated processes as collaborators with humans, not merely as simple replacements, opens up a whole new realm of possibilities for both humans and algorithms.
Around the same time, Alteryx hosted its first Twitter chat, addressing topics like the democratization of data and upskilling for the data professions. …