Businesses of all sizes can employ machine learning to power their applications and services. AI can work across a handful of areas, from automatic segmentation to fraud prevention chatbot services. But there’s a new kid on the block that’s stealing the show: AutoML (Automated Machine Learning).
Automated Machine Learning, you say?
AutoML is a new and quickly developing field with no strictly agreed upon scope or definition. Several companies such as Google are spearheading research and development of AutoML programs. A recent Google Research article explains that “the goal of automating machine learning is to develop techniques for computers to solve new machine learning problems automatically, without the need for human-machine learning experts to intervene on every new problem. If we’re ever going to have truly intelligent systems, this is a fundamental capability that we will need.”
Isn’t automation the point of ML to begin with? Why this new layer?
AI and ML algorithms require management from data scientists, engineers, and researchers—and there simply aren’t enough of those to go around. AutoML seeks to automate some of the repetitive, menial tasks that go into ML, such as choosing data sources, data prep, and feature selection. That way, data scientists can more efficiently take care of the involved processes that are harder to automate. Models can be built in less time and with more accuracy, allowing data scientists to spend more time fine-tuning newer algorithms.
Gartner estimates that by 2020, more than 40 percent of data science tasks will be automated. That’s created the new idea of “citizen data scientists,” allowing people who don’t have extensive education and training in data science or machine learning to operate and build complex models in a user-friendly environment.
Can you give me an example?
Sure can! Zylotech is a great example, in fact. We’re designed to automate the entire customer analytics process, with an embedded analytics engine (EAE) that features a whole host of automated ML models. The result is a clean, effortless process from data prep to unification; from feature engineering to model selection. We’re able to discover non-obvious patterns and insights in near real-time. This helps data scientists and citizen data scientists leverage their data to generate truly personalized 1:1 customer interactions.
Josh Fayer manages Marketing Communications at Zylotech, where he brings an extensive background in public relations and computer science. Josh Loves music! He sings with an a cappella group, Orange Appeal, (whose recent album Unpeeled he shamelessly plugs at work), and at home he loves to mess around with a variety of instruments. Josh enjoys spending quality time with his adorable dog, and exploring around his new home near the South Shore of Boston.
If you liked this post, check out our first post in a four-part series on why marketing executives should care about AI.