Skip to main content
Search
Audio Interview

Predictive analytics

Published on May 14, 2020   22 min

Other Talks in the Playlist: Interviews with business leaders and scholars

Please wait while the transcript is being prepared...
0:00
Interviewer: Dr. Siegel, thank you very much for sparing the time today. Let me go directly to the subject of this interview, and ask you to explain: what is it you do, how do you do it and why do you do it? Dr. Siegel: Sure, I've been in the field of machine learning and predictive analytics since the early 90s, first as an academic, and then since 2003 as an independent consultant, applying machine learning for business problems, running a conference series, and I wrote a book called 'Predictive Analytics'. Machine learning is when computers learn from examples and somehow find patterns or generalize from those examples, typically in order to make a prediction or some kind of discrimination about individual cases, so: 'are you going to buy this product, are you going to cancel your subscription, is there a picture of a traffic light somewhere within this image?', and for self-driving cars. The kinds of tasks that we want organizations and their machines to perform repeatedly, put probabilities on potential outcomes and behaviors of individuals or categories of images, sounds and speech recognition. Things that occur in great quantities and which we have a lot of data to learn from, in order to improve the ability for machines to do these things. That field's called 'machine learning' and when you're applying it to business problems, like with customer prediction, it's also often called 'predictive analytics'. Interviewer: Behind the words 'predictive analytics' I assume that there