As cities build upwards to accommodate growing populations, the safety of deep excavation, the process of digging large foundation pits to anchor high-rise buildings, has become a significant challenge in the construction industry. These pits must withstand the problem of shifting of the underlying earth, changes in groundwater pressure, and the heavy machinery while remaining stable enough to protect workers and nearby structures. Failures at this stage can trigger collapses, flooding or structural damage.

Work in the International Journal of Critical Infrastructures discusses an AI (artificial intelligence) system designed to improve safety monitoring at deep foundation pit support sites. The system aims to identify abnormal behaviour, such as unsafe actions, improper equipment use, or entry in restricted zones without protective gear, in close to real time so that warnings can be sounded in a timely manner.

Construction sites have traditionally relied on manual supervision and earlier generations of automated monitoring. But these approaches often struggle to detect unsafe behaviour quickly and accurately. Many systems record high false acceptance rates, meaning they mistakenly classify dangerous actions as safe. Others process video feeds too slowly to intervene effectively in rapidly changing environments.

The new system combines several advanced AI techniques to address those weaknesses. It begins by extracting key frames from surveillance footage using the fractional Fourier transform. This is a mathematical method that analyses data across different domains. By identifying the most informative frames rather than scanning every second of video, the system reduces computational load but still retains critical information.

The system then uses a spatiotemporal graph convolutional network, a form of deep learning that analyses both space and time data. The spatial analysis examines how workers and machinery are positioned relative to one another, while the temporal analysis tracks how movements change over time. Unlike conventional image-recognition models that treat frames in isolation, this approach captures sequences of actions and interactions. This is vital for working out what is happening moment to moment on the construction site.

The final step is to use a hybrid model that combines a convolutional neural network (CNN) with a so-called long short-term memory network (LSTM). The CNN can recognise visual features such as body posture or equipment shape. The LSTM can detect patterns in sequences of data. Working together, those two tools allow the system to determine not only what is happening in a single frame, but whether a series of movements constitutes a safety violation.

In their tests on active deep excavation sites, the researchers got a minimum false acceptance rate of 2.43 per cent and a peak abnormal behaviour recognition accuracy of 99.12 per cent. Processing time was as low as 0.19 seconds per analysis cycle, allowing near real-time monitoring.

A large-scale study published in theInternational Journal of Business Innovation and Research has looked at what factors lead to sustained gains in the construction industry. The team looked at 226 nationally registered firms and found that operational efficiency and collaboration, long seen as the sector's primary remedies for underperformance, are insufficient on their own to lead to sustained gains. Instead, the decisive factor is whether companies fundamentally rethink how they create, deliver, and capture value.

The research used a statistical tool known as Partial Least Squares Structural Equation Modelling to analyse information from the 226 companies and to look for any relationships between various organisational factors. The approach allowed them to look at how lean construction practices and strategic partnerships affect performance. It was also possible to discern whether business model innovation acts as a bridge between these strategies and measurable outcomes such as profitability, operational efficiency and competitive position.

Lean construction is a systematic project management approach designed to eliminate waste and maximise value throughout a project's lifecycle. Waste includes excess materials, redundant labour, delays, reworking, and poor coordination between contractors. Unlike simple cost-cutting, lean methods emphasise continuous improvement, integrated workflows, and delivering greater value to clients.

The study confirms that those companies that adopt lean practices do tend to perform better. However, the most significant improvements did not stem solely from streamlining their processes. Instead, lean thinking proved most powerful when it also prompted broader strategic change.

That broader shift is captured in the concept of business model innovation. A business model defines how a company creates value for customers, how it delivers that value, and how it generates revenue. Innovation in this context involves reconfiguring those core elements. For example, this might include moving from one-off, project-based contracts to long-term integrated service models, adopting digital coordination platforms, redesigning revenue structures, and embedding sustainability into what the company offers to clients.

Business model innovation was found to have a strong and direct positive effect on performance. More importantly, it amplified the impact of lean construction. When lean methods were embedded within a redesigned business model, performance gains were significantly greater than when lean was treated as a stand-alone efficiency tool. The research found that partnerships boosted performance only when it allowed companies to innovate in their business models. Access to shared knowledge, resources, and trust-based relationships yielded gains only if companies used them to reconfigure how they compete and deliver value.

In the age of global branding, instantaneous communication, and generative AI images, the symbols that we see in our daily lives circulate at an unprecedented rate. A study in the International Journal of Information and Communication Technology argues that if the symbols we share are to foster understanding rather than confusion, designers must treat them as carriers of cultural meaning, not mere decoration.

The team has used communication science, design theory, and semiotics, the study of signs and how they create meaning, to propose a systematic, evidence-based framework to identify, refine and test traditional cultural symbols. Their concept echoes an insight by Ferdinand de Saussure that suggests that a sign is not simply a form but a form bound to shared content. A flower or mythical creature, in this view, evokes memories, values and beliefs as much as it depicts the object it illustrates.

As digital platforms accelerate the circulation and mutation of images, we experience the fragmentation of symbols and signs. Moreover, in the age of generative artificial intelligence, almost all content is being cannibalised and regurgitated as derivative works, visual motifs are thus losing their inherited symbolism or, at best, being misappropriated or diluted. In the face of these changes, the researchers suggest that semiotics has now become a necessary part of creativity and perhaps the only hope of our conserving our symbols and their significance.

In their paper, the researchers discuss a five-step process beginning with systematic data collection and identification of culturally significant symbols. They followed this with a cross-cultural analysis, design refinement, and empirical testing. Statistical analysis together with expert review allowed them to look at specific symbols, such as the blue-and-white porcelain motifs featuring the lotus, peony, and plum blossom. As a good example of symbolic art, these patterns scored highly for clarity, adaptability, and perceived authenticity. The lotus is widely associated in East Asia with purity and renewal, the peony with prosperity and honour, and the plum blossom with resilience in adversity. Their visual simplicity combined with layered symbolism appears to aid translation into contemporary branding, the analysis found. More complex imagery failed to ignite the imagination of general audiences, although it was recognised as culturally significant by the experts.

Quantitative evaluation thus shows the different priorities associated with authenticity and meaning, challenging assumptions of universal interpretation for even familiar symbols that might be used in marketing and branding.

Mental health problems are among the most pressing of public health challenges, affecting millions across different age groups and societies. Depression, anxiety, and stress-related conditions rank among the leading causes of diminished quality of life worldwide. They exact a heavy social toll and economic cost. Yet diagnosis still relies largely on self-reported symptoms and intermittent clinical interviews, which means diagnosis is vulnerable to memory lapse, stigma, and limited access to trained professionals.

Research in the International Journal of Networking and Virtual Organisations discusses an artificial intelligence (AI) diagnostic system that can spot early signs of various mental health conditions by analysing how people write online. The model, known as a Fossa-based Graph Neural Network (FbGNN), examines language patterns in text drawn from social media platforms and online forums. Instead of relying solely on questionnaires, it studies sentiment-driven textual information, the emotional tone, word choices and behavioural cues embedded in a person's online writing.

The researchers explain that their system combines two advanced computational techniques. The first is the Fossa optimisation, a feature-selection method based on search strategies seen in nature. In machine learning, features are identifiable pieces of information, specific words, phrases or emotional markers. By applying Fossa optimisation, the system can filter out any irrelevant data from those features and identify pertinent indicators of mental distress.

The second component is a Graph Neural Network, a GNN. A GNN analyses relationships by representing information as a network of nodes and connections. Here, nodes correspond to features, and the connections are the interactions between them. This allows the model to detect complex patterns, such as recurring combinations of emotional expression and behavioural signals.

By training the system to classify text based on categories such as depression, anxiety, stress, bipolar disorder, suicidal ideation, and personality disorders, the team was able to then test its accuracy against known sample data. It was able to predict a person's mental health status with an accuracy of almost 99 per cent in the trials. Such accuracy would be useful in screening for mental health problems among a cohort of users, such as students, employees, or any other group. It would allow healthcare follow-ups to be directed at those most likely to have problems that might be addressed and would only miss one in a hundred. Further refinements of the system could bring that accuracy closer to 100 per cent.

Shobitha, G.S., Kataksham, V.S., Nagalaxmi, T., Spandana, V., Sreelatha, G. and Radha, V. (2025) 'A smart intelligent Internet of Things framework for predicting mental health', Int. J. Networking and Virtual Organisations, Vol. 33, No. 3, pp.251–278.
DOI: 10.1504/IJNVO.2025.151510

Digital payments are a routine part of daily life for many people. As such, the risk of online fraud is rising alongside this convenience. Identity theft, email compromise, scams, and misleading investment schemes all exploit technological weaknesses and often user naivety and can lead to big financial losses.

Research in the American Journal of Finance and Accounting has looked at technological threat avoidance theory (TTAT), a framework used to understand how individuals respond to technology-related risks. The study sheds new light on what motivates users to protect themselves from online financial threats, if they do at all. It considers user attitudes towards fraud and the perception of potential financial loss with the aim of identifying the specific influences that lead to a user taking protective action.

The team surveyed users of online payment platforms and found that rather than an abstract fear of fraud, the decisive factor in whether or not people took preventative measures was simply the perceived financial loss. This finding suggests that awareness campaigns focused on general threats may be less effective than approaches that point out the direct financial consequences of online fraud.

Online fraud costs us roughly US$1 trillion per annum, and it is likely that figure is rising year on year. There are millions of reported cases and probably many more that are never reported. The losses that people bear when a victim of online fraud erodes overall trust in the digital systems on which we rely. Moreover, widespread, organised fraud can disrupt financial infrastructure, threatening broader economic stability and making it almost impossible for regulators to maintain oversight and control.

Facing such problems, the digital economy needs technological innovation in payment systems to incorporate effective strategies to influence user behaviour. Such strategies need to make it difficult for users to compromise themselves through technological naivety. Policymakers, platform developers, and financial educators also need to help in the design of interventions that align perceived risk with actual behaviour and so strengthen the individual against threats as well as help maintain trust in digital financial systems.

The migration to electronic medical records, used by healthcare providers, hospitals, and medical insurers, continues. However, this switch from paper records is leading to an accumulation of data, a lot of which is in free-text form that cannot be processed easily by an algorithm searching for knowledge and looking for patterns.

A study in the International Journal of Business Process Integration and Management has looked at using basic text-mining methods to convert this free text, which might be as unsophisticated as the jottings of a doctor or nurse, into something more organised. This kind of processing could make decisions in medicine faster and more consistent as well as potentially opening up new avenues for medical research and epidemiology.

The research focused on the specific medical condition of lower back pain and the reports associated with it. Lower back pain is a big problem for a lot of people and a major reason people miss days in work or file for disability. Experts can evaluate symptoms and consider what medical scans show and make a diagnosis and offer a prognosis. Administrators have to read through reports manually to determine fees and payments. A system to convert free text to structured text would be a boon, allowing dates and diagnoses to be searched, checked, and analysed much more easily.

The team used pattern-matching rules to look for regular expressions that allow software to detect specific phrases or formats in text. This could then be used to extract clinical and administrative details. This rule-based text mining was combined with machine learning algorithms that can learn from past data and make predictions about new cases.

The researchers tested their system on 255 anonymised reports. Medical specialists validated the extracted information, confirming a precision rate of 98 per cent. The structured information was then used to train three established predictive models: AdaBoost, which combines multiple simple models to improve accuracy; Random Forest, which aggregates the results of many decision trees; and Support Vector Machines, which identify boundaries between categories in complex datasets.

In tests, AdaBoost achieved perfect accuracy in predicting when rest should be prescribed. Random Forest reached 91 per cent accuracy and 93 per cent recall, a measure of how many relevant cases are correctly identified, in return-to-work assessments. The Support Vector Machine recorded a 98 per cent recall rate in classifying disability cases.

Beyond performance metrics, the researchers argue that the approach reduces processing time and limits transcription errors. Because the extraction rules are explicit, the system remains interpretable. This is important, as decisions still need to be explained to patients and others regardless of how structured or unstructured the data is.

As self-driving, autonomous, vehicles head out on to public roads, one of the field's most persistent challenges remains collision avoidance in unpredictable traffic. A study in the International Journal of Vehicle Design discusses an artificial intelligence (AI) control system that has a 97 per cent success rate in avoiding obstacles, with a maximum response time of about half a second.

Urban roads present a shifting landscape of pedestrians, stalled vehicles, roadworks and erratic drivers. For a self-driving car, safe operation depends not only on accurate sensors but also on rapid decisions made under such uncertain conditions. Conventional obstacle-avoidance systems often rely on fixed rules or straightforward processing of sensor data. These approaches can sometimes fail in heavy rain, fog, or headlight glare.

Other systems that use reinforcement learning, a branch of AI in which the algorithm learns by trial and error, such as Deep Deterministic Policy Gradient, need a lot of computing power and often struggle to work quickly enough for real-world driving conditions.

The new approach described in IJVD builds on a reinforcement learning framework called Soft Actor-Critic, or SAC. In this computing system, the software actor proposes driving actions while the software critic evaluates whether or not the given manoeuvre would be sensible or not. SAC is designed to learn so that positive outcomes boost the actor-critic interactions that led to them. The system also embeds entropy, a statistical measure of randomness that allows it to continue to explore the best manoeuvres rather than settling prematurely on a single solution. This helps the system remain adaptable in uncertain environments.

The model also incorporates a self-organising cluster mechanism inspired by the collective movement of a flock of birds, that famously avoid mid-air collisions. At close range, a mathematically defined repulsion force pushes vehicles apart to prevent impact. At medium distances, a velocity calibration rule aligns speed with an ideal braking curve to reduce the risk of rear-end collisions. Additional rules govern wall and obstacle avoidance. This layered design allows multiple autonomous vehicles to coordinate their movements without relying on a single lead vehicle.

A study in the Journal of Business and Management has shown that self-esteem plays an important part in determining whether someone wishes to pursue a leadership role. The findings have implications for both organisational success and career development, underscoring, as they do, how self-esteem affects personal motivation.

The research suggests that self-esteem affects a person's regulatory focus, a psychological framework that influences how individuals approach challenges and goals. There are two main types of regulatory focus: promotion focus and prevention focus. Promotion focus is characterised by a drive for growth, achievement, and opportunity-seeking. In contrast, prevention focus is concerned more with the avoidance of failure, staying safe, and fulfilling one's basic duties and no more.

Individuals with high self-esteem are more likely to be promotion focused, which then drives them to seek leadership roles. Those with lower self-esteem tend to lean towards prevention focus, which makes them less inclined to pursue leadership roles.

The effect is not solely down to the individual's personality, however. The work also showed that career encouragement and support from supervisors and peers can affect a person's focus and the motivational pathways they might take. Encouragement can boost the positive effects of promotion focus, motivating individuals to pursue leadership. However, for those with lower self-esteem, encouragement can have the opposite effect, reinforcing their reluctance to take on leadership responsibilities due to their prevention focus. The research thus highlights a need to consider individual psychological states when offering career support so that talented people who have leadership potential are not lost to those roles because of their lower self-esteem.

The team adds that unlike static predictors, such as personality traits or gender, regulatory focus can be affected by one's experiences and external support. This makes it a more pliable characteristic that might be influenced to the person's benefit through good career development advice for those with the potential for leadership.

Research into the COVID-19 crisis, which began in December 2019, suggests that although there was widespread loss and disruption, the international crisis also planted the seeds for grassroots innovation and resilience. A study in the International Journal of Entrepreneurial Venturing of one hundred initiatives that emerged in Belgium during the pandemic finds that when established institutions struggled to respond quickly, individuals and organisations were able to step up to create new economic and social value.

The research focuses on initiatives defined broadly to include both newly created ventures and existing organisations that adapted their activities. These ranged from informal mutual aid efforts to repurposed businesses and newly launched services. Some were started by people with no prior experience of entrepreneurship. Other initiatives were started by established entrepreneurs responding to the sudden changes in demand and regulation. What they shared was a capacity to adjust rapidly under pressure.

The pandemic created conditions of extreme uncertainty. Lockdowns and business closures, imposed to limit the spread of the virus, caused sharp falls in income, consumption, and investment. Many people perceived formal support systems as too slow or rigid to meet urgent needs. This gap became the space in which these initiatives emerged, often spontaneously and with limited resources.

The study looks at this kind of resilience and rather than treating it simply as endurance in the face of a crisis, defines it as a dynamic process of recovery, adjustment, and innovation. Resilience was, during the pandemic and in its aftermath, both the route through which initiatives developed and the results they produced. The researchers argue that action was not driven solely by compassion or urgency, but by the ability to reframe the crisis as an opportunity to meet unmet needs.

The study suggests that locally driven, resilience-based initiatives can complement government and aid responses, particularly in the early stages of a crisis. As such, for policymakers, the challenge is how to recognise and sustain such efforts without undermining their flexibility. We will face pandemic and other shocks in the future, our ability to adapt and innovate in these conditions will be key to an effective disaster response.

A study in the International Journal of Global Environmental Issues has looked at "ritualistic" hunting practices in eastern India. It finds that they are contributing markedly to a worrying decline in wildlife and forest health. The work raises difficult questions about how cultural traditions can coexist with modern conservation goals.

The research focuses on Jungle Mahal, a forested region in western West Bengal, where hunting remains an integral part of religious and social life for several communities, particularly the Santhal. Ritualistic hunting, defined in the study as the killing of wild animals for ceremonial rather than commercial or subsistence purposes, is shown to be placing increasing pressure on ecosystems that are inherently vulnerable.

West Bengal hosts a range of ecologically significant species, including pangolins, fishing cats, and diverse bird populations. Such animals play crucial roles in the functioning of the ecosystems across the region. They help to regulate prey populations, disperse seeds, and recycle nutrients, among other things. The study reports a clear reduction in wildlife richness, biodiversity. It also notes a marked decline in forest density in Jungle Mahal. It is worth noting, that residents and hunters are well aware of these changes to their local environment, however, there is the paper reports, little inclinations towards matters of conservation.

Hunting in the region employs traditional techniques such as bow-and-arrow, traps, nets, and the use of smoke to flush animals from burrows. It occurs throughout the year, but intensifies during festival periods between March and June. During this period, large communal hunts with hundreds or even thousands of participants take place and huge numbers of animals are killed in a very short time.

India's Wildlife Protection Act of 1972 prohibits the hunting of wild animals, but the researchers found that enforcement is weak in remote forest areas. Awareness of conservation laws among local communities is limited, and illegal hunting continues unchecked. The study highlights the fact that there is great mistrust of authorities in such regions and a general perception that conservation policies are detrimental to indigenous values and livelihoods. It remains an open-ended question as to how this disconnection between culture and conservation might be remedied.