Artificial Intelligence
Summary
Artificial intelligence is a complex field and is being applied in a lot of different ways. We wanted to see if machine learning might help us and our clients to better anticipate customers needs. We also wanted to check out the state of Python and Javascript frameworks that implement machine learning algorithms. We’re still in-process, but are having fun learning how to predict customer churn and parse through huge sets of data.
The Objectives
- We wanted to learn about AI, specifically machine learning, to know what problems it might solve and to predict future outcomes.
- We wanted to tie this learning back to Ricochet’s work, to see how we might incorporate these algorithms into our services.
- Our ongoing goal is to have the expertise to advise and deliver solutions for our clients’ problems using machine learning.
Technical Details
Machine Learning is just one discipline of Artificial Intelligence - there are mathematicians and PHDs devoting endless hours working on this research, and as time passes their research filters down and is parsed into publically available libraries and algorithms. This, combined with the ever-increasing processing power of computers has made it easier than ever to learn and apply machine learning to practical use.
Though Python is a favored language for AI, we’re using our Javascript knowledge to evaluate and integrate libraries built to parse large data sets with the goal of predicting patterns in that data -- one problem we wanted to explore was predicting the point at which a customer might become dissatisfied with a service and leave. Using large data sets, we can evolve models to better predict these patterns before they happen. Our goal is to operationalize these technologies - to bring them to our clients and apply them to our work.
