A fascinating event organised in conjunction with the team in IBM Research, including presentations from leading experts focused on data privacy, machine learning and agile analytics in business.
David Braines, IBM CTO Emerging Technologies UK, presented on Fully Homomorphic Encryption. The common method of storing sensitive data and sharing it with colleagues and partners has a weak link. Today, files are often encrypted in transit and at rest, but decrypted while in use. This provides hackers repeated opportunities to steal unencrypted files. Homomorphic encryption plugs those holes.
David shared some applications of Agile Analytics used to forecast player performance in the Premier League.
Dr Maurice Coyle, Chief Data Scientist, Truata discussed some common myths in relation to Data Privacy and how to extract value from your data while protecting your customers’ privacy.
David Kearns, Product Manager, IBM Data Science Team, shared a presentation entitled Machine Learning Predictions With Natural Language Understanding: A Moneyball Story. How Machine Learning Models were used in the signing of a Major League Baseball player during the 2018 pre-season draft. The models created to predict pitcher and batter performances used cutting edge techniques like Auto-ML and derived attributes. A technique to illustrate how this information could be employed in any organization is demonstrated using the latest in Natural Language Technology.
IBM Research - Differential Privacy Library - Naoise Holohan, Research Scientist, IBM Research, provided an Introduction to Differential Privacy with Diffprivlib. The presentation will introduce the concept of differential privacy, and how it can be used to help manage the privacy risk of analytics on sensitive data.