Implementing AI Risk Management Frameworks
One way of doing this is for those organizations to prioritize the development and implementation of AI risk management frameworks and policies. The good new...
All Posts
2024
2023
Transforming Diabetes with AI - Temporal Fusion Transformers and Blood Glucose Data
Knowing where one’s blood glucose values have been and are right now is very helpful. What would be even more helpful is knowing where they will be in the fu...
Using LLM's For Private, In-House Document Analysis for Your Business
One of the main use cases that seems to be on everyone's mind is how to take advantage of LLM's by combining their natural language ability with your private...
Cloud LLaMa - Local LLM's and Getting Started with LLaMa on AWS EC2
In this blog post, we're goign to walk through running your own copy of a local, open LLM on the cloud (in this case AWS).
The Top 5 Analytics Challenges Most Organizations Face (and How to Solve For Them) - Challenge 3 - Solving for Compliance
A successful compliance practice starts with a well-designed data architecture, enabling the organization to manage and protect its data effectively.
2021
The Top 5 Analytics Challenges Most Organizations Face (and How to Solve For Them) - Challenge 4
I'm going to say something here that may strike you as a bit off, but it needs to be emphasized. Do NOT ask your customers what they want.
The Top 5 Analytics Challenges Most Organizations Face (and How to Solve For Them) - Challenge 5
Finally, and most importantly, let's talk about the critical piece that allows all of this work well
2020
ClickHouse, Redshift and 2.5 Billion Rows of Time Series Data
Let’s walk through using AWS to generate 2.5 billion rows of true time series data, and then using ClickHouse to demonstrate some very impressive performance.
2018
The 2018 Deep Dribble Bracket
Gonzaga. Syracuse. Mostly Gonzaga
2017
The 2017 Deep Dribble Bracket
This year we've got two brackets to work with.
2016
Deep Dribble 1.1 - And a personal update
I am now making predictions from Day 1
The 2016 Deep Dribble Brackets
Overall I'm happy with the first full 100% Python implemntation of the model.
Forecasting "The Walking Dead" Viewership via Time Series Regression with Seasonal ARIMA errors in R
I decided to try my hand and forecasting viewership for 'The Walking Dead'.
Evaluating Big Data Performance of PrestoDB and Parquet on S3 storage
As far as PrestoDB, I'm definitely impressed.
2015
It's Alive! DeepDribble 1.0 and 2015-2016 NCAA Basketball Predictions
I've christened the new model / interface DeepDribble, and it's currently live.
Amazon Web Services and Upgrading Python to 2.7 on EMR Cluster Installations.
EMR and the associated AMI's still included Python 2.6. I have no idea what that's all about. Maybe Jeff Bezos hates Python.
Trueskill And Nfl Football Team Ratings
If you’re not familiar with TrueSkill, it’s a rating system developed by Microsoft in order to assist in player matchmaking with their XBOX Live ...
Predictive Madness - Machine Learning and Modeling NCAA Basketball
It may turn out that boring is actually more correct, in which case I'll look silly, but it's all for fun anyway.
2014
Wordcloud Using R - North Korea's Press Release regarding Sony's "The Interview"
While we may not be able to watch The Interview any time soon..
The Dirtiest* Restaurants in Chicago (2014 Edition)
I decided I should probably not eat out so often...
A simple model for Kaggle Bike Sharing.
I've since developed a more accurate model, but I wanted to share a my initial attempt at a model for this problem that turned out very well. Simply repating...
LaTeX and Statistics Formulas
Since I'm ending up typing a lot of these in LaTeX lately, I'd figured I post them out there for reference.
Welch Free t-Test
When doing an independent two-sample t-test, the default for R is to assume that two variances are not equal, which is why it spits out a big line that says ...
Spurious Correlations
If you're looking for a good laugh, check out Tyler Vigen's website.
R Workshop
We had a great all-day R workshop on campus today, and I learned quite a bit.