Human augmented intelligence brings data to life
Professor Pablo Moscato’s work is expertly augmenting human intelligence with computational methods to produce an unprecedented transformation in the practice of decision making in Australia and around the globe.
Around the world, big data is radically changing the business and economic landscapes. Coupled with significant advances in artificial intelligence, machine learning and optimisation techniques, the way business is conducted is evolving at lightning speed.
Fortunately, data scientists like Professor Pablo Moscato and his team are here to help businesses, governments and industries navigate the future of innovation. With more than 30 years of research experience in a wide number of areas, he clearly articulates a vision of what’s to come.
“My research team uses advanced computer modelling and mathematical algorithms to detect patterns and predict outcomes of interest using the information in large datasets,” explains Professor Moscato.
“In today’s technology-focused world, enormous volumes of data can be generated on almost any topic—from marketing to biotechnology, water resource engineering to transportation. Data science allows businesses, governments and industries to harness the power of big data to solve complex problems.”
The power of memetic algorithms
Professor Moscato has always been a pioneer and champion of lateral thinking. Thirty years ago, he created an entirely new field of computer science known as “memetic algorithms”—labelled one of the greatest research frontiers in the combined fields of mathematics, computing and engineering.
Moscato explains the novel concept as a multi-algorithmic approach to solving problems, where single algorithms are replaced by smarter multi-algorithms. Based on a ‘survival of the fittest’ concept, his technique improves the quality of the algorithms and the solutions that they obtain for the problem of interest.
“Multi-algorithms outsmart single ones. In memetic algorithms, a set of autonomous computational agents act like a team to solve a problem. Instead of using a single algorithm to solve the problem, which is what’s happened in the past, we create a computational ecology that works together, killing off those that aren’t working and replacing them with ones that do.
“Our first memetic algorithm of 1988 included the idea of ‘battles’ and cloning the computer’s winner solutions. Solutions can then also be altered and ‘evolve’ inside the computer. Some of these ideas have become mainstream now. In the movie, Matrix Reloaded, the agents replicate themselves—it’s a similar idea.
“These computational ecosystems can help every industry to achieve their goals and overcome optimisation and decision-making challenges.”
Today, a Google search for memetic algorithms retrieves nearly 273,000 results. More than 22,100 academic papers cite the subject, and more than 700 papers published in China alone use the word “memetic” in their title. These impressive results make Moscato one of the world’s most cited computer scientists—and based on one of his earliest new ideas alone.
Understanding consumer behaviour
Professor Moscato’s work is having a resounding impact across multiple industries and fields, including marketing and business intelligence. Computer algorithms are creating smart data-driven methods of analysing online consumer behaviours and brand engagement. In the preface of his recently co-edited book, Business and Consumer Analytics: New Ideas (2019), together with his co-editor, he explains how marketers and decision-makers must adjust to this data-driven revolution.
“[The revolution] is fuelled by the increased availability of data gathered, and stored, by new technologies. Today, we are moving into the era of Data Science, and again, it all started not with products and services, but with the humble consumers and by putting them in the centre of the scene.”
Fuelled by the desire to create personalised solutions for consumers, Moscato recently proposed methodology for a new way of modelling human behaviour. Professor Moscato says the data-driven approach has the potential to reveal 'functional' relationships between the variables (i.e. actual interactions between measurement variables relating to behaviours), which can complement other pair-wise correlation studies of associations between variables. Results of the proposal were first published in the esteemed interdisciplinary journal, PLoS ONE, in July 2014.
“This methodology could be generalised and prove useful for future research in the fields of consumer behaviours using questionnaire data sets or studies investigating other types of human behaviours.”
Moscato’s methodology has already been used across multiple industries and fields, including the Australian not-for-profit sector. In 2015, his work helped Australian charities analyse their consumers’ behaviour through a survey conducted in partnership with the Australian Charities and Not-for-Profit Commission. The data allowed charities to proactively build consumer trust and loyalty by investigating the key drivers of charitable giving and understanding the values of their consumer groups.
In the non-profit and commercial sectors, Professor Moscato’s truly multi-disciplinary work clearly shows that enormous opportunity is now available to business and industry leaders looking to differentiate and better understand their consumers.
Revealing the changing patterns in literature
Moscato’s work has a multitude of surprisingly novel applications. In digital humanities, his data science expertise has provided valuable and fascinating insights into globally cherished artworks such as Shakespearean era plays and poems.
Using new algorithms he has created for the task, Professor Moscato was able to analyse works from the Shakespearean era to reveal authorship affinities. The project looked at word frequency profiles to uncover patterns of relationship between them, highlighting the connections with authorial canons.
Moscato and his peers found that “authors’ characteristic styles are very powerful factors in explaining the variation of word use, frequently transcending cross-cutting factors like the differences between tragedy and comedy, early and late works, and plays and poems”. Results were published in multiple journals, including the interdisciplinary and world’s largest journal, PLoS ONE, on several occasions.
The team’s innovative information theoretic clustering approach is now allowing other works of art to be examined in a similar fashion, providing an empirical guide to the authorship of plays and poems where this has previously been unknown or disputed.
Leading expert in personalised medicine
Curiously, but in what is second nature to him, Moscato brings innovation across what can sometimes be seen as highly dissimilar fields. For example, algorithms developed for the Shakespeare project motivated new applications in the field of personalised medicine, which involves targeting treatments for patients based on their genes, and other biotechnologies.
Moscato’s most recent medical research is exploring the mechanism of action for Alzheimer's and cancer drugs, which is largely determined in vitro, using data to generate new insights into complex health challenges.
Shortly after moving to Australia in September of 2002, Moscato established the Newcastle Bioinformatics Initiative (2002-2006), followed by the University’s highly interdisciplinary Priority Research Centre for Bioinformatics, Biomarker-Discovery and Information-based Medicine (2007-2015). He was also the leader of the Newcastle node of the ARC Centre of Excellence in Bioinformatics and had research contracts with the National Institutes of Aging (USA) to apply his skills to early detection of Alzheimer’s disease.
While his current work is even wider than the confines of his last centre’s boundaries, Professor Moscato maintains his conviction that computer science and novel biotechnologies hold a tantalising promise of allowing analysis of the molecular profiles of affected tumours, as well as normal cells. This could eventually lead to the automatic determination of which drugs are most effective for individual patients, rather than using the lengthy trial-and-error process.
Moscato says the approach will require the health system to establish itself as an adaptive, "learning from data" business intelligence operation. “Much more is needed than just collecting data,” he says.
“It's compatible to have such business intelligence with customised drug treatments at a personal level, identifying a way to both minimise patient dissatisfaction and reduce government costs. This way, pharmaceutical companies will also be given an indication as to where a drug is effective, so their next generation of medicines can be perfected to target specific problem areas.”
Revolutionising the future
Progress starts with data, but it goes much further than that. As we step into the future of technology, it’s becoming increasingly clear that capturing, interpreting and applying quality data is now key to driving innovation in every industry—which is what makes Moscato’s work so intensely valuable.
Moscato is convinced that the best translation comes from the most innovative new concepts in computer science, and good algorithms should be able to jump the field barriers and find new and much-needed applications elsewhere.
“The world is confronting us with unprecedented challenges, but we have never had so much data to best guide our decisions. We moved from the personal computer in the eighties, to the increasingly more familiar personalisation of services in this new century. With better decision making, all the sectors of the economy and life would benefit from methods that extract knowledge from data.”
Moscato is keen to continue collaborating with companies that will benefit from these disruptive changes, as well as the next generation of intrepid students that can pioneer these emerging novel methods.
“The future is bright, but we must make it fair and evenly distributed, as data analytics should serve us all.”