Hello, I'm Michael Hotaling

I'm a Data Scientist and Engineer at Applied Materials, putting the power of Artificial Intelligence and Machine Learning at the edge of Industry 4.0 and next-generation Smart Factory Automation.

I specialize in early failure detection for machinery, prognostic analysis, dashboards, visualization, and exploratory data analysis

I’m looking to collaborate on data science projects, machine learning, and applications of data science in other fields

Recent Work

Estimating Aqueous Solubility of Molecules

An exploratory view of how chemical solubility can be estimated using chemical attributes like molecular weight, aromatic proportions, and number of molecular bonds.

Modeling Industrial Machinery for Early Failure Detection

The OMCE Vega Shrink-wrapper is a machine heavily used in the food and beverage industry to group together loose cans and bottles into packages. A critical component is the shrink wrap cutting mechanism that, over time, becomes dull and may fail to cut.

Calculating Remaining Useful Lifetime for Turbofan Engines

An exploratory view of how chemical solubility can be estimated using chemical attributes like molecular weight, aromatic proportions, and several others.

Building a Hotel Recommendation System

Weekend getaway? Business trip? Expedia has a challenge for data scientists to contextualize customer data and predict the likelihood a user will stay at 100 different hotel groups.

AI Generated Cookbook

Wouldn't it be nice to let a robot make dinner for you? Here is a compilation of recipes generated by our artificial intelligence system. Bon Appétit!

Designing an Unbeatable Poker AI using Reinforcement Learning

Poker is one of the quintessential games that involves making decision based off of imperfect information. For this project, we will explore the process of developing a program that can best humans in a game of Texas Hold’em.

Using Statistics and Machine Learning to Predict Winning Horses

By analyzing physical attributes of horses such as average speed, number historical wins, terrain conditions, weather, and dozens of other data points, we can create a model that can return probabilities of certain horses winning.

Predicting and Preventing Auto Loan Defaults

This case study describes the analysis of auto loan defaults from a particular bank. The objective of the case study is to determine which of our clients are more at risk for loan default using several machine learning algorithms. By using these techniques, we can mitigate the risk of defaults by determining which customers are more likely to pay of their debt.

Detecting Insider Trading Using Anomaly Detection

By looking at investor volume and historical transactions, we can identify unusual patterns in activity and detect insider trading patterns.

Recreating the Moneyball Model

Using data analytics and moneyball theory, Oakland A's general manager, Billy Beane, hired the best players he could with an extremely limited budget for payroll.

• Full Portfolio