For Chris Roth, programming always came natural. “I’ve been programming for ages. I started when I was around 10 years old, so I’m one of those people.” His first computer was a Gateway PC running Microsoft Windows, providing Chris with exposure to technology during the dawn of the consumer Internet. “I was learning HTML, a little bit of JavaScript, and C++. So I started programming because I was trying to modify video games that I was playing online.”

It became apparent to Chris that the skills that he was developing had the potential to make him employable. At 20, Chris left North Carolina State University to work as a software engineer in several start-ups — he also performed his share of freelance work. His first full-time gig landed him at a media analytics company out of Colorado. There, he performed real-time video and audio demographic analytics to provide live broadcasters a way to gauge viewership in real-time. His next job brought him to New York’s, an online education platform focused on software engineering, business, and marketing.

With that experience behind him, Chris returned to freelancing and working on his own projects. With a desire to focus on data science, he began working on Chronist, a utility that uses another open source project, Lifeslice, to capture snapshots of your face at regular intervals. “My end goal with Chronist was to see if I could come up with some sort of way to quantify emotion over time in an objective way. The main way people assess their emotions is to answer a series of questions.” Chris wanted to show that emotion measurement could move beyond a questionnaire and into the realm of active analysis.

With Affectiva’s JavaScript SDK and some programming, Chris found that the visual data that Chronist captured could be easily analyzed over time, providing insights into one’s emotional state over a longitudal period of months and even years. In addition to videos, Chris analyzed text. “The period I analyzed depended on the stream of information, so I had eight years of Facebook messages, but I only had one year of the photos.” His friend, Stan James, had even more photo data to analyze — six years worth! So what did Chris discover with Chronist? “I’m still kind of working on figuring things out. I’m working with a couple of other people that have strong math backgrounds to see they can ascertain anything.”

“There are a lot of patterns in the world, and any insights that you can gain from this data can range from obvious to subtle. Emotion analysis is an example of that… trying to take slight variations in someone’s expression over a long period of time, and tease out information from it.” With data science tools like Tensorflow and Spark, Chris believes that it’s easier than ever to jump into the field. “It just seems to be a very practical and interesting area to move into.”

So what’s next for Chris and Chronist? “For now, the goal isto put this in front of people that see its potential,or that have really strong data science backgrounds. I think there’s probably a lot more you can do that we haven’t done yet.”