Uncovering Team Performance Dynamics with Data & Analytics in Complex Engineering Projects

This talk introduces a framework by the Global Teamwork Lab (GTL) at the University of Tokyo and MIT to uncover the nature of performance during complex projects. The most innovative and significant grand challenges for industry and society are marked by technical and social complexity, with teams working across boundaries.  With recent capabilities to instrument demands and activities, we propose a new lens and inquiry into the performance of teams.  Sensors on both the people and the problem are analyzed in real-time, so that the awareness, interaction, and actions by teams are enhanced.  An integrated “meso-scale sociotechnical systems” approach requires integrated instrumentation, analytics, modeling, and visualization so that data is streamed, processed, considered, and acted upon in the cognitive sweet spot of human teams.  We’ll show some recent experiments from GTL and the new “Interactive Visualization Lab” at MIT.

Anomaly Detection with Deep Learning: Finding the Needle in the Haystack in the Enterprise

Deep Learning is taking research by storm with applications in ad technology, image recognition,self driving cars, and speech processing. In this talk we will talk about another area of application for deep learning: anomaly detection. Anomaly detection includes areas such as: preventative maintenance in factories, network intrusion, and fraud detection. In this talk we will cover why deep learning is suitable for this area of application and two approaches for doing anomaly detection with deep learning.

We will finish with an end-to-end production case study to show how a complete AI stack can be built for this use case.

AI for Supply Chain

Solving practical business decision-making problems using predictive analytics involves much more than just data visualization or applying traditional forecasting techniques. Inclusion of domain knowledge, advanced forecasting techniques and intuition can create a successful model that avoids catastrophic failure. I’ll dive into one best practice learned from solving real world problems in the area of supply chain management.

Most supply chain management companies are frustrated with inaccurate forecasts from customers and ERP systems. They are constantly searching for more reliable forecasts so that they can dramatically slash inventory costs while providing better, more efficient service to customers.

Drawing inspiration from the applications of artificial intelligence and from mathematical problem solving skills, I will walk through some examples that illustrate how creative and innovative the solutions are, when problems are solved with AI and big data.

Data Science Initiatives at a FinTech Company

Within the FinTech industry, which has gained attention in recent years, what kind of services does Money Forward provide? This talk will introduce how Money Forward considers such initiatives as data science and machine learning, and how it promotes them as an organization. While there are many elegant examples of initiatives externally, this talk will introduce how to overcome unavoidable obstacles as new initiatives are advanced, including struggles and gotchas to watch out for.

Big Data Analysis for Cyber Security

Recently, the leaking of personal information due to cyber attacks has become never ending. Because the methods of attacks have become more sophisticated, it is difficult to prevent them. When a security incident occurs, it is necessary to localize the damage through rapid detection and response. In this presentation, I will introduce visualization and methods for effectively detecting attacks using big data, in this case, huge log data. Along with introducing cyber attack methods, I will suggest how to use big data for cyber security by analyzing logs.
Participants will learn about:

  • Methods of attack, such as targeted attacks.
  • How to use logs to detect attacks.