BC Data Colloquium: Andrew Henrey, Phat Tran& Dmitry Samosseiko

We conclude the annual BC Data Colloquium with a three part discussion series on May 17. More details on the demonstration and  talks are available below.

Phat Tran,

Head of Shared Services IT,Technical Safety BC

Computing on edge and real time object detection Demo

Bio: Long time IT pro and a tinker at heart. Using YOLO with transfer learning for fast object detection at the edge. Let me set it up and show it to you and you can help me draft the intro of this demo.

Dmitry Samosseiko,

Director of Global Threat Research at Sophos.

Redefining Cybersecurity with Data Science

The machine learning technology took the cybersecurity industry by storm a couple of years ago. Nowadays, nearly every cybersecurity product claims to contain some form of “artificial intelligence”. In this talk we’ll look at why and how the IT security became such a rich soil for AI application, what problems it was set to solve and how much progress has been made so far.

Bio:Dmitry Samosseiko is the Director of Global Threat Research at Sophos. He is leading a large team of cybersecurity experts at SophosLabs and is driving strategy and innovation in the areas of anti-malware, threat intelligence and messaging security. Earlier in his career, Dmitry was a software developer at ActiveState Inc., where he played a key role in founding its email security product line, which led to company’s acquisition by Sophos in 2003. At that time, Dmitry ran a SophosLabs team in Canada and was responsible for local efforts in threat research, protection development and automation. With over 15 years of experience in IT security, Dmitry has been a frequent speaker at industry events, published research papers and has been quoted by the media on topics of cybercrime and cybersecurity.

Andrew Henrey,

Data Scientist, Finning Digital

Analytics Solutions in the Heavy Equipment Industry

Today's connected world enables countless opportunities for innovation in the heavy equipment industry. For Finning, a leading heavy equipment dealer, this means new digital solutions for optimizing maintenance plans, reducing inventory costs, maximizing machine uptime, improving safety and several other avenues to deliver value to our customers. The key to innovation is a diverse set of operational and transactional data. Every machine transmits massive amounts of operational data every day and all opportunities, decisions, and sales are logged for future analysis. There are new sources of information, including drones providing real-time geospatial information and wearables providing real-time health and safety information about personnel. This provides a rich environment for data scientists to develop predictive and prescriptive solutions that address key business questions: How likely is a system failure in the immediate future? What is the industry demand forecast for heavy equipment? How can we create business opportunities from our connected assets? Our talk will focus on analytics solutions that we have developed, and are developing, to answer these problems.

Bio: Andrew's experience lies in developing and implementing cutting edge machine learning solutions for video game analytics, arthritis research, and industrial experiments. At Finning Digital, Andrew is primarily focused on implementing the newest methodology in business forecasting. Andrew holds a PhD in Statistics from Simon Fraser University, with a particular focus on Random Forests to model changing variance. When Andrew isn't working, you can probably find him playing video games or classic board games.