What to do About Super-Charged AI and Machine Learning

Artificial intelligence and machine learning is all around us today from recommendation engines that serve up internet ads to conducting fraud checks as we process our digital payments.

But not all machine learning/artificial intelligence is the same: some is built with offline models whereas others are built with real-time infrastructure.

DBTA recently held a webinar with Brian Bulkowski, CTO & founder, Aerospike, and Matt Bushell, director of product marketing, Aerospike who discussed the key characteristics between offline and online AI and machine learning, defined “hungry data” and more.

Machine learning is rapidly growing and IDC forecasts that spending on AI and ML will grow from $12B in 2017 to $57.6B by 2021.

Machine learning can offer companies:

  • Churn prevention
  • Demand forecasting
  • Price optimization
  • Quality assurance
  • Up- and cross-selling
  • Customer lifetime value
  • Fraud detection
  • Predictive maintenance
  • Risk management
  • Customer segmentation
  • Next best action
  • Product propensity
  • Text mining

A variety of industries put AI/ML to good use including financial services, pharmaceutical, healthcare, insurance, manufacturing, oil and gas, internet and technology, media and retail, and banking, according to Bulkowski and Bushell.

Machine learning works best when companies have a well defined, specific goal, understand the business problems and rules, have lots of data, and have the ability to train the data.

“Hungry AI” is result-oriented AI through online learning, automatic optimization, and automated feature selection, Bulkowski and Bushell explained.

An archived on-demand replay of this webinar is available here.