William Stevenson: Dude, where's my data analyst? (remotely from Boston)

Dude, where's my data analyst? A quick guide to using machine learning in a real-world environment by William Stevenson (https://metacpan.org/author/WDS).

You wake up in the morning to a refrigerator full of data, no memory of what happened the night before and no idea where the person is who you hired to analyze the information. What do you do? This talk should give you a map of how to use machine learning in a real-world environment and a sense of how to navigate that territory.

I work as software engineer and data analyst at MaxMind. Recently I have helped develop a platform to predict the riskiness of various components of an e-commerce transaction. This code will touch hundreds of millions
of transactions per year with sub-second response time; however, I was a Philosophy major and have no formal training in either data analysis or computer science.

I took a Massive Open Online Course (MOOC) on machine learning, but found that it focused too much on the details of particular algorithms and required a heavy background in math and statistics. I also found a disconnect between what I was learning and how to apply it at work.

Instead I worked through the chapters in Machine Learning with R and, after understanding how to use each algorithm, applied it to the problem I was trying to solve. From this emerged a few final candidates and also a deeper understanding in myself of how to use machine learning in the real world.

After integrating one algorithm into production, I gave a tech talk to my coworkers. One of them then encouraged me to give this presentation to others.

In having to teach this to myself, I hope I'm in a good place to teach it to you.

(If you're interested in some reading material to prepare for this meetup, you may want to read Machine Learning with R (https://www.packtpub.com/big-data-and-business-intelligence/machine-learning-r).)