We are proud to present several BDVA’17 Data to Decisions CRC researcher talks.

Niall Fay

Chief Operating Officer, D2D CRC

Presentation Title

Big Data for national security: finding the needle in the data haystack

Presentation Synopsis

The threat landscape for national security agencies continues to evolve at an unprecedented rate. Terrorist and criminal activities are growing globally and becoming increasingly sophisticated. The use of new technologies by these groups offers them the ability to plan and coordinate attacks and crimes with impunity. The sheer volume of data and complexity driven by the use of diverse software applications means much of the activity remains hidden. The need to develop and apply advanced data analytics capabilities and tools has never been acuter and it applies across a broad range of national security domains, including border security, financial intelligence, defence, counterterrorism and cybersecurity. This presentation will describe some of the new capabilities and tools that must be developed to support the national security agencies to achieve their mission of creating a safer and more secure nation.

Lewis Mitchell

Lecturer, University of Adelaide

Lewis is a lecturer in applied mathematics at the University of Adelaide. His research focuses on large-scale methods for extracting useful information from online social networks, and on statistical techniques for inference and prediction using these data. He works on building tools for real-time estimation of social phenomena such as happiness from written text and assimilating observational data in predictive mathematical models for weather or disease prediction. Perhaps unsurprisingly given his research interests, Lewis is an avid Twitter user who you can follow at @lewis_math.

Presentation Title

Predicting civil unrest from open data—from research to product.

Presentation Synopsis

In this presentation, Lewis will discuss from a research perspective the experience of working with D2D CRC and its end users on a project to predict civil unrest events from open data sources. This project takes real-time data inputs and applies statistical and machine learning algorithms to filter, clean, and then make predictions from these large, unstructured datasets. The team consists of a diverse set of researchers from applied mathematics, statistics, and computer science, and involves personnel from postdoctoral fellows and researchers, to software engineers, and members of the industry. The talk will outline the research problem and discuss some of the approaches that have been taken to solve it. It will also describe the team’s experience in attempting to turn research outcomes into a product for end users.

Duane Rivett

Commercialisation Manager, D2D CRC

Presentation Title

Fivecast—Taking the product to market

Presentation Synopsis

Translating research into commercial success is an inexact science. What is the key ingredient? If it’s great science, surely that’s enough. Brenton will discuss the approach taken within the D2D CRC that has led to Fivecast—a startup with a bright future and in doing so uncover the “secret sauce”.