Artificial Intelligence Skills and Education

As the sun rises over Sydney’s iconic Opera House, another day dawns for Australia’s job market that rapidly navigates the AI revolution. From Woolies self-checkouts to algorithmic AFL outcome predictions, from automating routine tasks to complex problem solving, AI tools have long transformed the nature of work, job descriptions and skill sets we see in demand across the nation/s labour market. More than that, AI advancement moves faster than a kangaroo can jump. To future-proof Australia’s work needs, we must understand how our education and training can meet these skill needs to stay ahead of the game. 

Recent research insights at University of Sydney and Macquarie University revealed the top Australian job titles/roles that most likely require AI skills. We used a proprietary data set of 2.7 million job ads from 2016-2024 to identify the job roles requesting AI skills. AI skill taxonomies are in development, we applied the AI skill classification as defined by Lightcast in the OECD 2023 report. It clusters AI skills in artificial intelligence, autonomous driving, machine learning, natural language processing, neural networks, robotics, and visual image recognition.  

Top 15 jobs that demand AI skills down under 

The heavy hitters in the AI skill arena down under are: 

Software Developers 
Computer Systems Engineers/Architects 
Computer and Information Research Scientists 
Management Analysts 
Market Research Analysts and Marketing Specialists 
Information Technology Project Managers 
Data Scientists 
Network and Computer Systems Administrators 
Information Security Analysts 
10 Marketing Managers 
11 Sales Representatives (Wholesale and Manufacturing, Technical and Scientific Products) 
12 Sales Managers 
13 Medical Scientists (except Epidemiologists) 
14 Computer and Information Systems Managers 
15 Computer Users Support Specialists 

It’s no shocker that technical roles are at the top of the list when it comes to AI skills—they naturally need heaps of IT know-how. This 2022 report by the Australian government and National Skill Commission found that the AI surge created new job titles like Data Scientist and Data Engineer that do not even exist in the Australian occupational classification system. But what’s really interesting is the way AI skills are creeping into not-so-technical fields like Marketing and Sales. Jobs in these areas are now using AI for things like analysing social media sentiment, optimising marketing campaigns, creating customer recommendation systems, and building chatbots for customer support. 

Even more fascinating is the rise of managerial roles needing AI skills. Managers use AI for data-driven decision-making, evaluating organisational performance, improving hiring and training programs, and streamlining communication. This shows how AI is becoming integral to all areas of business. Finally, a bit further down the list, we see scientists jumping into the AI game, too, which means AI skills are spreading across the job market, well beyond just IT roles. 

You might notice generative AI skills—like those involving ChatGPT—aren’t on the list. That’s because these technologies only became mainstream in early 2023, while Lightcast set their definitions in December 2022. But expect to see skills like prompt engineering, ethical use of AI tools, and model training and tuning to become more common, pushing even more non-technical jobs to the top of the AI job lists. 

But it’s not just about technical skills 

Building national AI capabilities isn’t just about tech savviness. AI skills require higher-order thinking skills like critical thinking and creativity. Long gone are the days when it was enough to be able to write code, develop an algorithm or be proficient in prompt engineering. Practitioners need to understand the wider implications of AI integration to to ensure AI practice is ethical, inclusive and equitable, serves the common good and doesn’t hurt individuals. For instance, HR managers need to think before they deploy AI in recruitment understanding inherent biases and ensuring the technology promotes fair and inclusive hiring practices. 

Case in point is our current research on the impact a company’s AI integration has on managerial roles. Our preliminary findings point to the changing of manager skills and the increasing number of various skills as organisations (we look at F100 companies) integrate AI. New manager skills in demand have nothing to do with AI. These skills include organising and planning work tasks, coordinating integration across various organisational functions, decision-making abilities, agile thinking, etc. 

Role of higher education 

This is where Australian universities shine and rise to the challenge. They offer safe spaces, social connections and multidisciplinary ecosystems to develop critical thinking, ethical awareness, drive for justice, and a mindset for social good. University is where future generations are nurtured and responsible citizens are grown to advocate for themselves, others and the planet. That’s what universities do. Australia’s higher education is the nation’s third largest export and is a 47.8$ billion dollar industry, which suggest we grow quite a substantial share of the future generation worldwide.  

Like other universities worldwide are grappling with how to respond to national AI future skill needs, Australian universities also need to keep up with changing demands for upskilling, not only in terms of content but also format of education. Here are some solid game plans for how universities can tackle these challenges: 

  1. Interdisciplinary Programs:  The future is interconnected, we can expect maths to shake hands with music, and science swapping notes with art! We are likely to see a spur of programs mixing, blending and creating new discipline areas. Interdisciplinary courses should connect technical AI learning with ethics, humanities, and social science so students get the full picture of complexity. 
  1. Liberal Curriculum: For Undergraduate students specifically, a broader liberal  curriculum could focus on the holistic student development that hones in on those uniquely human skills. Think courses like interpersonal skills, genuine listening, caring and advocating for others, ethics and justice. These can give students greater purpose beyond the first entry level job, and prep them for head-scratching ethical dilemmas when it comes to AI. 
  1. Learning Partnerships: Not only the disciplines should be the ones talking to each other, universities for boost the authenticity and relevance of their education by partnering up with tech companies and startups to give students the real life practice experience and the networks they want. Many universities already offer real life project based work in capstone courses before releasing the graduates into the wild wild workplace. These need to be resourced to pack a punch and generate real impact! 
  1. AI or Tech labs or incubators: Let’s finally build those bridges between academy and industry where knowledge generation meets application in real time, where ideas can be born slowly, but things can be built and broken rapidly, in order to advance our thinking. These labs can be Australia’s innovation hotspots, bring knowledge alive, facilitate crowdsourced innovations, run hackathons open to public, and empower Australia’s talent. 
  1. Lifelong Learning: With the AI advancement travelling at the speed of light (aka much faster than a kangaroo) we need to develop a mindset for continuous learning. We cannot afford to stop learning at graduation. Think short sprints, online modules, and certified credentials to keep with the workforce needs. It’s not only good for the economy, it also boosts your personal growth and pumps up your confidence muscles. 

Higher education in Australia redefines what education we provide and, perhaps even more importantly, to whom and how. 45% of Australians aged between 25-34 has some higher education degree and many look to upskilling and continuous learning to advance professionally and grow personally. However, they are all busy, juggling numerous work and life responsibilities and commitments, so delivering flexible education options to suit individual needs is key. 

Skills needed to do a PhD: recent research using text mining and machine learning

The number of people with PhDs is growing worldwide. We know that doing a PhD is a significant undertaking and dropping out of one can result in serious financial loss, psychological issues such as loss of confidence, etc. 

It is not surprising that a lot of research exists on the doctoral experience with the aim to improve it and to find out what it takes to finish it. 

A body of research also exists that looks at the outcomes of a PhD, i.e. what do people gain from a PhD? This question is gaining importance because there seems to be an oversupply of PhD graduates for academia, which means PhD holders need to seek jobs elsewhere, I.e. in corporate or public sectors. 

We analysed PhD requirements to find out what PhD students need in terms of skills, attributes and qualifications.

We analysed the selection criteria for PhD candidates on a platform that advertises PhD programs as job ads. Our analysis of thousands of these ads revealed exactly what types of skills different countries and disciplines require. See the infographic below for a quick summary of the findings and implications of this research:

This study draws on the data source of PhD role advertisements (aka ‘PhD ads’) to identify what skills and/or other requirements doctoral programs seek before PhD admission. We analysed the selection criteria of 13,562 PhD ads posted in 2016-2019 on EURAXESS – Researchers in Motion, a pan-European initiative by the European Commission.

We developed a taxonomy based on the EURODOC ‘Transferable Skills for Early-Career Researchers’ framework and extracted attributes present in each advertisement. To do this we employed text mining and machine learning approaches. We created an updated taxonomy using data-derived dictionary. 

You may use the interactive dashboard below to search for details on PhD ads in any of the 50+ countries and any of the 30+ disciplines represented in the data sample (2016-2021).

Dashboard with data up to 2021

See the full paper for full details on the methodology and the research study:  

Note: The paper is based on 2016-2019 data only and the sample for this time period can be accessed here.

Read The Conversation article on this research.

Read the Campus Morning Mail post on this research.

Entity linking Systems for Literature Reviews

Let’s face it. In busy academic lives there is hardly any time to do some deep reading, let alone stay up to date with everything that is published in your area in real time. Yet, staying up to date with latest knowledge and reviewing literature regularly is our bread and butter as academics.

In addition, published literature reviews help establish your expertise of a particular area. Researchers increasingly automate the coding process in literature reviews and accelerate the literature review process by using computer-assisted tools like Leximancer, topic modelling, Bibliometrix, R packages, NVivo, etc.

However, existing approaches for coding textual data do not account for lexical ambiguity; that is, instances in which individual words have multiple meanings.

To counter this, we developed a method to conduct rapid and comprehensive analyses of diverse literature types by using entity linking in literature reviews. We present a new literature review framework that embeds entity linking.

See the framework step by step below:

In the same paper, we present an example where we apply the framework to review the literature on digital disruption and digital transformation.

On how to adapt the framework to your needs, see the full paper:

Marrone, M., Lemke, S., Kolbe, L.M. (2022), Entity linking Systems for Literature Reviews, Scientometrics. Forthcoming.,

Trends in FinTech Research and Practice: a systematic review

Many industry sectors have experienced significant disruption in recent years through the introduction of new financial technology (or FinTech), including process automation in financial services and the adoption of cryptocurrencies. From the first telegraph cable in 1866 to blockchain in 2009, the evolution of financial technologies has always been aligned with innovations in information systems (IS).

How do FinTech and Information Systems relate to each other? Where are the crossovers, where do they intersect, where do they diverge?
This question drove me and my colleagues to conduct a systematic literature review and to compare academic with practitioner literature.

Findings from our review show that the practitioner-oriented literature foreshadowed the rise of FinTech by extensively reporting on algorithm-based and electronic trading (2009 onwards), followed by reporting on FinTech start-ups and funding successes (2014 onwards).

The practitioner literature subsequently reported on alternative finance models, the introduction of cryptocurrencies, and risks and regulatory issues. Academic literature on FinTech began to rise from 2014 onwards, focusing initially on the development of FinTech in the aftermath of the 2007-2008 global financial crisis.

Research attention subsequently shifted to FinTech innovations (alternative finance, cryptocurrency and blockchain, machine-based methods for financial analysis and forecasting, including artificial intelligence), as well as risk and regulatory issues.

IS work on FinTech started to emerge from 2015 onwards, initially focusing on mobile payment systems and peer-to-peer lending. However, the body of work at the intersection of FinTech and IS is still small.

Changes in FinTech literature over time

Our review sheds light on several opportunities for future research, including financial inclusion, the impacts arising from COVID-19, and the emergence of new business models, such as Banking as a Service (BaaS).

Full paper reference: 

Cai, C., Marrone, M., & Linnenluecke, M. (2022). Trends in FinTech Research and Practice: Examining the Intersection with the Information Systems Field. Communications of the Association for Information Systemshttps://www.researchgate.net/publication/359107231_Trends_in_FinTech_Research_and_Practice_Examining_the_Intersection_with_the_Information_Systems_Field

Developing interdisciplinary research maps from business/management and the environmental sciences

Summary of “Interdisciplinary Research Maps: A new technique for visualizing research topics”(Marrone & Linnenluecke, 2020)

Interdisciplinary research is challenging, in part due to the sheer magnitude of knowledge embedded within disciplines, and also to the lack of a common shared understanding across them.

To bridge the gap in understanding between the disciplines of business/management and the environmental sciences, Marrone & Linnenluecke (2020) developed a ‘map’ of topics, concepts, and ideas discussed in top publications in these fields of research.

Developing the map

  • Data for the study was sourced by selecting articles published since 2011 in the top four journals by impact factor in each field, through the Scopus database. The abstracts, titles and publication years of 4,827 environmental sciences articles and 2,671 business and management articles were downloaded.
  • These data were exported to two separate Comma Separated Values (CSV) files, one for each of the areas of interest. The titles of the publications were then merged with their respective abstracts and the files were analysed using TAGME entity linking tool to compile a list of all possible topics from the text in the abstracts and titles.
  • The researchers ‘cleaned’ the results of the analysis by deleting topics that made little meaningful sense, given the context in which they were used. After cleaning the results, 7,915 topics were retained in the environmental sciences articles, and 4,293 in business/management articles.
  • A map was created (see figure below) to show the frequency with which topics are mentioned in each field.

What the map tells us

1.     Topics that are frequently identified in one literature, but not the other

Some topics are represented almost exclusively in the environmental sciences articles, many of them linked to concerns about climate change. Meanwhile, in the business/management journals, the topics which arise most frequently are related to firm structure and expansion. There is an opportunity here, for future research to connect topics across these two fields. For example, further interdisciplinary research could serve to explore the impacts of climate change on business and management decisions, such as asset valuations and investments.

2.     Topics that are frequently associated with both literatures

Several topics are common to both sets of literature. For example, “decision-making” is a frequently discussed topic in both business/management and the environmental sciences. The topic of “China” also arises frequently in both disciplines, but in different ways. Articles in the business/management journals address management challenges and economic opportunities in China, while the those in the environmental sciences journals address the role of emerging economies like China’s in climate adaptation and mitigation efforts. Such areas of topic convergence may provide fruitful avenues for future interdisciplinary research.

Figure 1: Topics are represented as dots; those associated with the business/management literature are coloured red, and those associated with the environmental sciences blue.

Reference:

Marrone, M. and Linnenluecke, M.K., 2020. Interdisciplinary Research Maps: A new technique for visualizing research topics. Plos one, 15(11), p.e0242283.

Tracking trends in environmental accounting research using machine learning

Summary of “Trends in environmental accounting research within and outside of the accounting discipline” (Marrone, Linnenluecke, Richardson & Smith, 2020)

Environmental sustainability concerns us all. Its importance is reflected in the exponential rise in the profile of research into accounting for environmental degradation, which has taken place since the establishment of the 1997 Kyoto Agreement.

To identify those areas of environmental accounting research which might benefit from a greater exchange of ideas between accounting and non-accounting disciplines, Marrone et al. (2020) utilised a literature review powered by machine learning. The review tracks the emergence of topics and trends, both within and outside of the discipline of accounting.

The review process

  1. A range of keywords were applied to a Scopus database search, which returned 2,502 records. Eighty-three percent of these were published in non-accounting journals.
  2. The TAGME Entity linking system was used to extract topics within the titles and abstracts of these journal papers.
  3. A burst algorithm was then applied. This identified ‘hot’ topics, looking at publications over time– ‘bursts’ indicate new developments related to the topic or a sudden surge of publications in a topic area.

The findings

The review compared two bodies of literature. The figures below show trending topics over the past 50 years, in accounting and non-accounting journals. Those that were trending in 2019 are highlighted in red.

Comparison shows that research in the field of accounting has recently focused on the connection of environmental accounting with corporate social responsibility (CSR) and stakeholder theory. But outside of the accounting journals, more specialised sustainability topics are explored. These include the shift to a low-carbon or circular economy, the attainment of sustainability goals (SDGs) and newer concepts such as accounting for ecosystem services.

Figure 1 Timeline of accounting journal bursts.

Figure 2 Timeline of non-accounting journal bursts.

One reason for the difference between the bodies of literature could be that accounting research is turning away from practical, interdisciplinary issues in favour of building on the theoretical foundations of the discipline. An increased exchange of ideas across disciplines could both strengthen the theoretical basis of research published in non-accounting journals and increase the range of emerging sustainability topics explored in accounting journals.

In future, the method of this review may be developed further, allowing for a more fine-grained analysis which produces updates as new issues of journals are released. Additionally, a means of quantitively examining topic exchanges and cross citations could enable more accurate comparison of the relative relationships between literature reviews. Further improvements in natural language processing may also facilitate an increase in the quality of the automated coding conducted by entity linking tools.

Intelligent Machines: Will AI replace academic researchers?

According to a recent article in Forbes magazine,we can expect to see most people collaborating on their work with an AI (Artificial Intelligence) counterpart by 2030. Will these ‘counterparts’ simply enhance what we do, or will they ultimately come to replace human beings altogether?

Transhumanism

Transhumanists examine the possibilities presented by the interactions between people and technology. They tend to welcome technological developments as means of enhancing our intellectual, physical, and psychological capacities (Bostrom, 2003; figure 1).

Figure 1. Adapted from Bostrom (2003, p.12)

An ethical minefield?

It’s not all good news, however. The concept of merging people with technology throws up a host of ethical considerations.

  • Who will access technological advantages? Those who can afford to pay for AI implants, wearable technology and tools? Will a group of enhanced humans emerge at the expense of everyone else?
  • If a person performs a criminal act, while coupled with an AI element, who bears responsibility? Should it be the person linked to the AI, the AI itself, or the person who developed or programmed the AI?

How close are we to developing human-like intelligence?

The idea of people integrating with computers is far from new. The famous computer pioneer J.C.R. Licklider coined the term ¨man-computer symbiosis” in the 1960s. Licklider predicted that computing technology would eventually advance to the extent that you would be able “to think in interaction with a computer in the same way that you think with a colleague whose competence supplements your own.” (Licklider, quoted in Lesh et al., 2004). Almost 60 years on, are we there yet?

The sort of technology which could result in human-like intelligence and the sorts of ethical conundrums outlined above, is still some way off, yet there are some very interesting possibilities available to us. Researchers, for example, may approach a form of symbiosis with computers. AI-based search tools are developing, which can help us navigate the broad swathes of literature available to us, speeding up our engagement with information (Extance, 2018). This is where we believe ResGap comes in.

Harnessing the potential of AI

Technology such as that available through ResGap can enhance your research performance, by allowing you to quickly find research gaps, understand how your research field has evolved over time, and identify hot and cold topics. By letting ResGap do the filtering for you, pointing you to where you will most productively direct your attention, you are able to engage with an astonishing breadth of information.

The kind of technology offered by ResGap doesn’t replace you as a researcher. You can work withResGap, but not yet in Licklider’s “Man-computer symbiosis”. Rather, you are still in the driving seat, harnessing the power of an extremely useful tool. While you won’t be able to access the whole space available to posthumans in Bostrom’s model (above: The Space of Possible Modes of Being), but you’ll certainly be pushing at the envelope of what is accessible to humans. It’s not a bad place to start.

Bostrom, N. (2003) Introduction to transhumanism. Presented at the Intensive Seminar on Transhumanism, Yale University, 26 June 2003. Available at:  https://www.slideshare.net/danila/introduction-to-transhumanism?from_action=save

Extance, A. (2018) How AI technology can tame the scientific literature. Nature Available at: https://www.nature.com/articles/d41586-018-06617-5

Lesh, N., Marks, J., Rich, C., Sidner, C.L. (2004) Man-Computer Symbiosis ´Revisited: Achieving Natural Communication and Collaboration with Computers. IEICE Transactions on Information and Systems

A tale of two research fields: from Team Mental Model theory to research model creation

Team Mental Model Theory

Team mental models are organized mental representations of the team’s relevant environment, shared across team members (Klimoski & Mohammed, 1994). They emerge because individual team members tend to categorise elements of their environments, such as tasks, situations, response patterns or relationships. These categorisations then become shared over time, thanks to communication within the team.

The extent to which mental categorisations are shared across team members can vary widely. They may be highly consistent with one another or completely incongruent. Importantly, when researchers talk about shared mental models, they do not suggest that an identical set of categorisations is held by every member of the team. Rather it is suggested that there exists some degree of consistency or convergence between individuals’ mental models (Kang, 2006; Rentsch, 2008).

Teams with higher levels of convergence of mental models can perform better. A shared mental model enables team members to anticipate the needs and actions of others in the team (Cannon-Bowers, 1993). Thus, the team can coordinate its actions, enhancing its decision-making capacity (Stout, 1999).

From TMM to research models

If we conceptualise research communities as teams, we can begin to see how TMM theory applies to researchers seeking to identify what is researched by individuals members of the team. A degree of sharedness will exist across the mental models held by researchers in each field. They are likely to use the same terms in reference to the same concepts. However, mental models which are shared within one field may diverge significantly from those in a neighbouring field. An understanding of the ways in which different groups think about a topic has vast potential in the pursuit of interdisciplinary research.

Our map, generated by resgap.com help illustrate the idea that mental models lie on a continuum, rather than as a dichotomy (say, very infrequently discussed in one literature to smilingly identical mentions in both literatures). As an example, we could study how the term “techno-stress” is studied by Psychology and Information Systems researchers. Our tool would help uncover what are the topics that are frequently discussed in both fields, as well as topics that are frequently discussed in one field, but unfrequently discussed in the other.

By mapping what is discussed among researchers in separate fields, we may increase the effectiveness of our research. It is possible to develop a systemic understanding of who is doing what, increasing the coordination of our actions with those of other members of the research community. We become able to anticipate the needs and actions of our fellow researchers.

The unique contribution of resgap.com lies in its use of entity linking. Because resgap.com does not rely on keyword identification alone, it identifies discussion of concepts, not just the usage of similar terms. This helps overcome the challenge presented differences in the use of terminology. Different terms may be used in different fields to describe the same concept, or the same terms may be used in different fields to describe completely different concepts. However, resgap.com can cut through the confusion, allowing us to see where fields overlap, and where relevant and valuable research may be directed.

References

Cannon-Bowers, J. A., Salas, E., & Converse, S. (1993). Shared mental models in expert team decision making. In N. J. Castellan, Jr. (Ed.), Individual and group decision making: Current issues(pp. 221-246). Hillsdale, NJ, US: Lawrence Erlbaum Associates, Inc.

Kang, H.-R., Yang, H.-D., & Rowley, C. (2006). Factors in team effectiveness: Cognitive and demographic similarities of software development team members. Human Relations, 59(12), 1681–1710. https://doi.org/10.1177/0018726706072891

Klimoski, R., & Mohammed, S. (1994). Team Mental Model: Construct or Metaphor? Journal of Management, 20(2), 403–437. https://doi.org/10.1177/014920639402000206

Marrone, M, Hammerle, M (2018) Smart Cities: A Review and Analysis of Stakeholders’ Literature. Bus Inf Syst Eng60: 197. https://doi.org/10.1007/s12599-018-0535-3

Rentsch, J. R., Small, E. E. & Hanges, P. J. (2008) Cognitions in organizations and teams: What is the meaning of cognitive similarity? In D. B. Smith (Ed.),LEA’s organization and management series. The people make the place: Dynamic linkages between individuals and organizations (pp. 127-155). New York, NY, : Taylor & Francis Group/Lawrence Erlbaum Associates.

Stout,R.J., Cannon-Bowers, J. A., Salas, E., & Milanovich, D. M. (1999). Planning, Shared Mental Models, and Coordinated Performance: An Empirical Link Is Established. Human Factors, 41(1), 61–71. https://doi.org/10.1518/001872099779577273