Category: Data Science

Miroslav Rajter builds a Monty Hall problem simulator using R:

The original and most simple scenario of the Monty Hall problem is this: You are in a prize contest and in front of you there are three doors (A, B and C). Behind one of the doors is a prize (Car), while behind others is a loss (Goat). You first choose a door (let’s say door A). The contest host then opens another door behind which is a goat (let’s say door B), and then he ask you will you stay behind your original choice or will you switch the door. The question behind this is what is the better strategy?

This is something that puzzled me for a very long time. This is fundamentally a Bayesian problem built around processing new information, and once I understood that, the answer was a lot clearer. H/T R-Bloggers.

Comments closed

Allison Koenecke takes us through the arulesSequences package in R:

In the following tutorial, we answer both questions using the R package arulesSequences [4], which implements the SPADE algorithm [5]. Concretely, given data in an Excel spreadsheet containing historical customer service purchase data, we produce two separate Excel sheet deliverables: a list of service bundles, and a set of temporal rules showing how service bundles evolve over time.  We will focus on interpreting the latter result by showing how to use temporal rules in making predictive sales recommendations.

Our running example below is inspired by the need for Microsoft’s Azure Services salespeople to suggest which additional products to recommend to customers, given the customers’ current cloud product consumption services mix.  We’d like to know, for instance, if customers who have implemented web services also purchase web analytics within the next month.  Actual Azure Service names have been removed for confidentiality reasons.

Market basket analysis is an interesting topic, though in my limited experience, it really falls apart when you have a large number of products to compare, so it tends to work better with toy examples or limited product selections because when you have a 50,000+ SKU inventory, the lift of any individual combination of products rarely gets above the level of noise.

Comments closed

I wrap up my series on the Naive Bayes class of algorithms, finally writing some code along the way:

Now we’re going to look at movie reviews and predict whether a movie review is a positive or a negative review based on its words. If you want to play along at home, grab the data set, which is under 3MB zipped in 2000 reviews in total.

Unike last time, I’m going to break this out into sections with commentary in between. If you want the full script with notebook, check out the GitHub repo I put together for this talk.

Assuming I ever get a chance to do this talk again, I’m probably going to change the data sets in the example given how overplayed iris is.

Comments closed