Joining us today we have my esteemed colleague Paul Spiegelhalter. Paul is a data scientist and machine learning specialist with expertise in predictive analytics and algorithmic modeling across a number of industries, including computer vision, online advertising and user analysis, medical diagnostics, natural language processing, and anomaly detection. There’s a lot of hype in the media these days around Machine Learning, Data Science and Artificial Intelligence. Today we are going to discuss the difference between the terms, really concentrating on what machine learning is at the 101 level and we want to talk about how it applies to you in the enterprise. As always we will also include some handy tips on how you can get started and how you can learn more about it.
Key Points From This Episode:
- How Paul got into machine learning and data science.
- Machine learning as a way computers recognize patterns in things.
- The different forms of Artificial Intelligence.
- Data science: the intersection of machine learning, domain knowledge and problem solving.
- Find out how machine learning works through feeding information to algorithms.
- Two different forms of machine learning: supervised and unsupervised.
- Steps to take for a machine learning project.
- Doing machine learning on different platforms.
- Good examples of machine learning projects – facial recognition.
- How machine learning can do limited strategy adjustment.
- What sequence to sequence learning is.
- Identifying machine learning opportunities.
- Knowing when to get a machine to do the programming.
- The importance of sentiment analysis.
- Decent findings in the outliers.
- Understanding transfer learning.
- The benefits of having some level of technical training and statistical knowledge when getting involved with machine learning.
- Visual recognition, a project Paul has been the most proud of.
Links Mentioned in Today’s Episode