Digital innovation: Delivering magic through collaborations

In today’s evolving world, businesses recognise the need to transform, innovate, and adopt new tools and technology to stay competitive.

The University of Auckland’s School of Computer Science is an institution that prides itself on developing digital innovation solutions to support businesses in their endeavours to solve real-world problems.

The School of Computer Science supports research that aims to discover novel and viable digital innovation solutions with a wide range of industrial applications.

Through its partnerships with industries and businesses, the school has developed digital solutions that range from measuring emotions in a cinema to developing a computer-aided assessment of spinal pathologies.

Building smart chatbots capable of working out complex solutions

Chatbots play a pivotal role in digital automation, streamlining interaction between customers and services, and provides an endless array of applications spanning various industries.

Current chatbot technology is limited by its inability to reason, only allowing it to answer very simple questions that can be extracted from predefined text keyed into its knowledge base.

Michael Witbrock, a professor at the University of Auckland’s School of Computer Science is currently working on developing systems that could pave the way for the invention of smart chatbots that are capable of solving complex queries.

Professor Witbrock’s research aims to develop computer systems that can learn to reason, thus enabling it to use that reasoning to put knowledge together and ultimately solves complex problems.

“The ultimate goal is to build a system that recursively decomposes questions, extracting answers for the decomposed questions and assembling answers, using salient background knowledge,” Professor Witbrock said.

Although the research approach will enhance the capabilities of chatbots, it can be applied more generally to any question-answer system, the potential industrial applications of these findings are vast.

“Having systems that understand even some of the vast flood of knowledge and data we produce and can assemble that information with the attention and speed and breadth of reach computers can achieve, can have positive effects in almost any industry, “ he added.

Before joining the School of Computer Science, Professor Witbrock was a Distinguished Research Staff Member at IBM T J Watson Research Center in Yorktown Heights, New York.

The research team led by Professor Witbrock is currently developing active partnerships with researchers at Massey, Waikato, and Otago Universities.

The research team also has ongoing research collaborations with both the AI for Good Foundation (AI4Good.org), and IBM Research located in New York, Boston, and Zurich.

Measuring human emotional response to movies, by analysing trace gases found in breath

Joerg Wicker, a senior lecturer at the university’s School of Computer Science, along with his colleagues at the school discovered a way to tangibly assay the human response to emotional events.

Wicker’s research findings sought to understand the human body’s responses in terms of exhaled air to external emotional stimuli.

Traditionally, emotional stimuli were carried out through EEG or skin resistance techniques that are often obtrusive. Instead, this new research proposes to measure Volatile Organic Compounds (VOCs), that can be detected at quite low concentrations in exhaled air in response to emotional stimuli.

This discovery has introduced a new field of application for data mining where elements found in trace gases released as a response by people reacting to films shown in cinemas are related to the semantic content of the films themselves.

These findings have paved the way for the development of non-invasive techniques to measure emotions with a wide array of applications across sectors ranging from psychology, health, elder care, to commercial applications, for example, measuring response to new movies or video games.

Studying changes in bat population to determine water quality

Joerg Wicker and Yun Sing Koh, senior lecturers at the university’s School of Computer Science collaborated with a group of environmentalists to learn how changes in bat populations can be used as a proxy for measuring water quality.

This research aims to find a cost-effective system that can be used to measure environmental pollution in hard-to-reach bodies of water based on bat calls.

This study was made possible through a collaboration between the university, Waitakere Trust, a non-profit organisation, and the Auckland Council.

The techniques applied in the research involve the use of using sensors to record noise at the frequency of bat calls, automatically identifying bat calls, and estimating and monitoring the population of bats.

Developing a computer-aided assessment of spinal pathologies from computed tomography (CT) images 

The spine forms the main support system of the human musculoskeletal system and plays a major role in mobility. Despite its importance, diseases and injuries commonly affecting the spine, such as osteoporosis are still often underreported.

Dr Martin Urschler, from the School of Computer Science, University of Auckland, in collaboration with Dr Darko Stern and Christian Payer from the Institute of Computer Graphics and Vision, Graz University of Technology, Austria set out to develop a computer-aided assessment of spinal pathologies based on computed tomography (CT) images as an aid for spinal experts.

The study aims to develop an accurate method for detecting and delineating the spine vertebrae in CT images.

The method developed through this study aims to be robust enough to be effective even in the presence of pathologies such as fractures, or in the presence of implants.

This discovery results in a wide array of potential applications that range from the automatic detection and assessment, or diagnosis of spine and vertebrae related pathologies to the development of an automated garbage sorting system!

If you are looking for a smart solution to solve a problem in your business, learn more about the School of Computer Science’s digital innovation capabilities by visiting their website.