Research in the International Journal of Computational Science and Engineering has developed a new approach to addressing ideological polarisation on social media. The problem of users generally encountering only like-minded perspectives and so reinforcing their own beliefs even in the face of conflicting evidence is highly divisive.

The phenomenon, known as the "echo chamber" effect or referred to as "filter bubbles", arises in part because the algorithms driving the position of content in one's social media apps. This, in turn, is driven largely by the need to keep users active and engaged on a particular platform. Too many contrary updates might drive users away, and that will ultimately reflect negatively on the advertising and other revenue streams for the companies that operate the platforms. By contrast, an echo chamber effect that reinforces their viewpoints will, for many people, be more attractive than one that doesn't.

Zaka Ul Mustafa and Muhammad Amir of the International Islamic University Islamabad, Manal Mustafa of Zaman Technologies Pvt Limited, Pakistan, and Muhammad Adnan Anwar of Ulisboa, Portugal, suggest that the social media platforms could benefit from the use of genetic algorithms (GAs). Such computational techniques inspired by the principles of evolutionary natural selection could reduce polarisation and the echo chamber effect but still respect the organic nature of online interactions, and so keep users engaged without being so divisive.

The team explains that current strategies to counter polarisation often involve connecting disparate groups (edge addition) or altering expressed views (opinion flipping). These methods are not only static, but also raise ethical concerns about the platforms interfering with user autonomy. A GA-based approach instead identifies influential nodes in the online social network and only subtly adjusts their highlighted connections to reduce polarisation. The critical contribution of the work lies in identifying network elements that disproportionately contribute to ideological divides, and then encouraging more diversity of interaction with minimal disruption to the organic nature of social media.

The team has tested their approach on real-world datasets that focus on polarised US political discourse. The datasets have communities clustered around distinct ideological groups, and so can provide a useful test for how well the method precludes polarisation and division. The results showed that the GA approach could foster connections between disparate groups, and this led to a measurable decrease in polarisation without fundamentally altering the network's overall structure.

Ul Mustafa, Z., Amir, M., Mustafa, M. and Anwar, M.A. (2025) 'Harmony amidst division: leveraging genetic algorithms to counteract polarisation in online platforms', Int. J. Computational Science and Engineering, Vol. 28, No. 7, pp.1–17.
DOI: 10.1504/IJCSE.2025.143956

As international trade and global security become more reliant on marine resources, the demand for advanced maritime surveillance and port management has never been greater. One of the big challenges in this area is the detection of ships in complex environments, a task that has traditionally relied on manual techniques. These methods, while functional, are often inadequate in dynamic, cluttered marine conditions, where varying sea states, weather patterns, and ship sizes can easily confound detection efforts.

Research in the International Journal of Information and Communication Technology has introduced a new approach to ship target detection. The research combines several cutting-edge deep learning techniques, "You Only Look Once" version 4 (YOLOv4), the Convolutional Block Attention Module (CBAM), and the transformer mechanism. The team of Weiping Zhou, Shuai Huang, and Qinjun Luo of Jiangxi Polytechnic University in JiuJiang, and Lisha Yu of Shanghai Cric information Technology Co. Ltd. In Shanghai, China, have combined these into a single algorithmic program that is both accurate and reliable in the identification of vessels in challenging conditions.

Modern, fast deep-learning models such as YOLOv4 out-class traditional methods by cutting out the multiple steps needed to process an image. YOLOv4 can scan and classify objects in a single pass, making it ideal for real-time surveillance over large expanses.

CBAM is a feature-enhancing technique that works by focusing the model's attention on the most important elements within a given image. This allows the hybrid system to identify ships even if they are surrounded by other vessels, docks, flotsam, and even rough seas. Conventional techniques often failed in distinguishing vessel from background in such images. The transformer mechanism is a powerful system that further improves the capacity of the model to process features at different levels, ensuring that important detail are not missed.

The team explains that this combined effort allows their system to outperform earlier models, particularly in the detection of smaller vessels and ships in complex maritime environments. They tested the approach on the Ship Sea Detection Dataset (SSDD), which includes remote sensing images of various marine conditions. Their results demonstrated superior speed and precision, especially when identifying minor or obscured targets. Given the critical importance of timely and accurate detection in maritime security, the implications of this improvement are significant.

Zhou, W., Huang, S., Luo, Q. and Yu, L. (2024) 'Research on a ship target detection method in remote sensing images at sea', Int. J. Information and Communication Technology, Vol. 25, No. 12, pp.29–45.
DOI: 10.1504/IJICT.2024.143631

Architects and industrial designers play an important part in what we might term the circular economy (CE). This is a sustainability framework that aims to minimize waste by reusing and regenerating resources. Research in the Journal of Design Research has surveyed practitioners in The Netherlands and Sweden to see whether there is growing enthusiasm for circular design strategies and what significant challenges remain to be overcome.

Giliam Dokter, Jonathan Edgardo Cohen, Sofie Hagejärd, Oskar Rexfelt, and Liane Thuvander of Chalmers University of Technology, Gothenburg, Sweden, surveyed 114 professionals. They found that almost two-thirds of them engaged with CE-related projects, while a similar proportion reported that there were shifts within their organizations to support such initiatives.

The team reports that techniques such as "design for disassembly", the crafting products or buildings for easy dismantling and reuse, are all part of this move towards greater sustainability. They point out that circular business models, emphasize regeneration over consumption and the associated principles are commonly applied in CE-focused projects undertaken by the survey participants.

It was found that architects tend to prioritize material reuse at the building level, while industrial designers have more of a focus on making it possible to disassemble products. Both groups are advancing creative solutions that reflect the principles of CE, however, even if their approaches are different and the substantial barriers they face are apparent.

The survey revealed that a lack of reliable knowledge about materials and the tools needed to evaluate environmental and economic impacts during design is one of the biggest barriers to adopting the principles of the CE in both architecture and industrial design. The research points out that choosing sustainable materials requires precise data about the lifecycle of these materials and their potential reuse. However, such information is often scarce or fragmented.

In addition to this dearth of relevant information there are also factors such as regulatory and market challenges that are beyond the immediate control of those working to CE principles and such barriers might hamper their efforts towards sustainability regardless of their efforts and focus.

Dokter, G., Cohen, J.E., Hagejärd, S., Rexfelt, O. and Thuvander, L. (2024) 'Mapping the practice of circular design: a survey study with industrial designers and architects in the Netherlands and Sweden', J. Design Research, Vol. 21, Nos. 3/4, pp.177–209.
DOI: 10.1504/JDR.2024.143685

Online shopping in China, particularly among young people, is a vast enterprise. Online retail sales amounted to about 16 trillion yuan in 2024, approximately 2 trillion US dollars. Indeed, online shopping has transformed the way youngsters approach buying everything from clothing to gadgets, especially in the post-pandemic era where old shopping habits have been abandoned by many people.

Much of the research into online consumer behaviour has focused on the after-sales experience. Now, a study in the International Journal of Data Science, turns the research lens to look more closely at the pre-purchase stage. In so doing, Nanhua Duan and Jingwen Zhang of Northwestern Polytechnical University in Shaanxi, China, hoped to understand how young Chinese consumers perceive value before they hit the all-important "buy now" button when shopping online.

The team explains that the concept of Customer Perceived Value (CPV) is at the core of their research. CPV refers to the overall worth a consumer assigns to a product based on the benefits they expect in relation to the cost. For experiential products, this perception is even more complex because the product's value is influenced by a variety of factors that may not be immediately obvious. The same is true for clothing when one cannot touch or try on an item before making a buying decision.

To home in on the factors involved, the team has proposed a new framework, which identifies six key dimensions that influence CPV when young Chinese consumers shop online for clothing and similar items. These are: word-of-mouth value, service value, aesthetic value, cost value, quality value, and brand value. Each of these, they found, plays a critical role in shaping the consumer's expectations prior to purchase.

The findings are particularly relevant to China's booming apparel market, which has seen rapid growth among digitally consumers. The research emphasizes that young buyers are not just concerned with the price tag or material quality alone. Indeed, they also consider factors like the reputation of the brand, the service experience, and how well a product aligns with their personal style or social status. This is where the online shopping environment differs from traditional brick-and-mortar shops, where the tactile nature of the shopping experience provides more immediate and obvious feedback and the potential for impulse buys or purchases prompted by an enthusiastic sales assistant.

For retailers and brands looking to tap into the ever-growing online market, understanding the six dimensions of CPV could offer insight into how to develop a more compelling online experience. It is, the research suggests, no longer sufficient to highlight the physical attributes of a product, companies must also now showcase the brand and its reputation as well as the quality of service.

In practical terms, the findings could mean that companies could benefit from focusing on positive reviews, clear and appealing product images, and smooth, customer-friendly websites. There might even be potential for developing innovative ways to display the products that might involve interactive elements, such as changing viewing angles, product colours and styles, and perhaps even offering options to see different models wearing the items. There is huge potential for the marketers that learn how to persuade people to click that "buy now" button.

Duan, N. and Zhang, J. (2025) 'The development of a product-layer perceived value scale for the online experience products of young Chinese consumers: take online apparel as an example', Int. J. Data Science, Vol. 10, No. 5, pp.1–21.
DOI: 10.1504/IJDS.2025.143886

Many work-related activities come with a risk of musculoskeletal problems, not least working at a desk. They are perhaps more commonly seen in the industrial or manual labour settings where repetitive movements, awkward postures, considerable muscular force and vibration, and lifting heavy objects are problematic.

A new study in the International Journal of Human Factors and Ergonomics introduces a tool that could be used by employers to assess the risk of such problems to their workers. The tool, the Ergonomist Assistant for Evaluation (ERAIVA), could streamline the process of identifying risky postures, which might lead to chronic pain and issues such as repetitive strain injury over time.

Where workers perform tasks that involve awkward body positions, repetitive movements, and heavy lifting there is an increased risk of debilitating conditions such as back pain and injury, carpal tunnel syndrome, and tendinitis. Previously, assessing such risks was done only on an ad hoc basis and not necessarily systematically, to the detriment of workers moreover the assessment itself was labour and time intensive, requiring experts to visually monitor workers or examine video footage of their activities.

Veeresh Elango, Lars Hanson, and Anna Syberfeldt of the University of Skövde, Staffan Hedelin and Johan Sandblad of Scania CV AB in Södertälje, and Mikael Forsman of the KTH Royal Institute of Technology in Stockholm, Sweden, explain that ERAIVA addresses these shortcomings by offering an automated way to analyse and annotate video recordings of industrial tasks. The technology could avoid human error in assessing work tasks and the posture and activity of individuals carrying out those tasks. Such a system could allow posture and other problems to be corrected and reduce the risk of musculoskeletal problems.

The system is easy to use and so reduces the need for expert assessment and remediation. Engineers and operators, as well as risk assessors, can all work together with the results it provides to identify and mitigate risks in the workplace.

Elango, V., Hedelin, S., Hanson, L., Sandblad, J., Syberfeldt, A. and Forsman, M. (2024) 'Evaluating ERAIVA – a software for video-based awkward posture identification', Int. J. Human Factors and Ergonomics, Vol. 11, No. 6, pp.1–16.
DOI: 10.1504/IJHFE.2024.143861

Online education is now ubiquitous and in recent years has changed fundamentally the way many people learn. Various platforms have opened up access to knowledge for millions of people. However, there remains an ongoing challenge: how to accurately measure and enhance the quality of teaching in these digital spaces.

Conventional evaluation tools focus on test scores and student satisfaction surveys. However, these often overlook the students' emotional experience of the course. Research in the International Journal of Information and Communication Technology, proposes a new solution that could change the way online teaching is assessed, getting closer to the heart of emotional matters.

The new work by Ruiting Bai of Puyang Medical College in Puyang, China, introduces the EduSent-Dig model, which can carry out advanced sentiment analysis and use big data techniques to evaluate teaching quality. By analysing the student emotional response given in their course feedback, the model can extract the nuances of online teaching that work most effectively. Rather than flagging the feedback as simply "positive" or "negative", EduSent-Dig identifies specific emotional undercurrents such as joy, frustration, or surprise. It does so by using analytical tools such as Bi-LSTM, a deep learning framework, and Word2Vec, which converts words into numerical representations for computational analysis.

The study reveals that emotional experiences are not just peripheral to learning; they are central to it. How students feel about their coursework directly affects their motivation, engagement, and whether they complete a course. As such, the new model in identifying and interpreting sentiment accurately, can provide educators and course designers with insights into how to improve their educational offering. Moreover, real-time sentiment analysis undertaken as a course progresses might even allow teachers to fine tune their teaching dynamically, tailoring lessons to student needs on an ad hoc basis. This could transform the way courses are designed and how they are developed as the students progress through them. All in, the insights could foster a more empathetic and effective learning environment.

Bai, R. (2024) 'Big data-driven deep mining of online teaching assessment data under affective factor conditions', Int. J. Information and Communication Technology, Vol. 25, No. 11, pp.35–51.
DOI: 10.1504/IJICT.2024.143412

Increasing complexity, evolving consumer expectations, and tightened development timelines means that physical product development increasingly comes unstuck when conventional methodologies are used. The predominant systems engineering frameworks have structure and predictability, but often falter when innovation is needed to fill the gap in modern markets. Companies have turned to agile approaches to help them transform their approach to software development, for instance. But, there are major obstacles to the adoption of that kind of approach for the development of physical products, where material constraints, prototyping costs, and supply chain integration are always critical factors.

A new hybrid framework is discussed in the Journal of Design Research that might address some of the issues. Frank Koppenhagen, Tobias Held, of Hamburg University of Applied Sciences in Hamburg, Tim Blümel of Porsche AG in Weissach, Paul D. Kollmer of the University of Hamburg, Germany, and Christoph H. Wecht of the New Design University in St. Pölten, Austria, describe a new model, Systematic Engineering-Design-Thinking (SEDT). In this approach, the strengths of systems engineering is combined with the user-centric, principles of design thinking to create a more adaptive and innovative product development pathway. SEDT builds on the Stanford University ME310 process, which has proven itself to some degree in academia and industry, but an expansion was always needed.

By integrating systematic exploration techniques from systems engineering, SEDT refines the ME310 framework to better support the development of solutions to problems. The result is a process capable of accommodating greater degrees of uncertainty and complexity, enabling teams to pursue transformative innovation rather than simply incremental improvement. The approach reimagines project structures to emphasize collaboration, fluidity, and cross-disciplinary interaction.

The next step is to test SEDT in both academic and industrial environments to determining its usefulness as a comprehensive framework for physical product innovation.

Koppenhagen, F., Blümel, T., Held, T., Wecht, C.H. and Kollmer, P.D. (2024) 'Hybrid development of physical products based on systems engineering and design thinking: towards a new process model', J. Design Research, Vol. 21, Nos. 3/4, pp.210–261.
DOI: 10.1504/JDR.2024.143686

Research in the International Journal of Information and Communication Technology suggests that machine learning tools might be used to detect and so combat financial fraud.

According to Weiyi Chen of the Monitoring and Audit Department of the Financial Shared Center at the National Energy Group Qinghai Electric Power Co., Ltd. In Xining, China, financial fraud is a constant challenge for capital markets, especially in developing economies where regulatory systems are still not fully mature. Fraudsters use sophisticated techniques to outpace conventional detection methods, which can leave investors exposed to potentially devastating risks beyond the everyday risks of investments! Chen's work offers a promising new approach to fraud detection by combining machine learning and deep learning to bridge the gap between financial data and the information found in corporate reports.

Financial fraud has long afflicted markets, distorted investment decisions, and weakened public trust in financial systems. Manual audits and statistical models can detect some fraudulent activities, but they can be inefficient when faced with increasingly complex fraud in the digital age. The problem is especially obvious in developing markets, including China, where financial fraud is widespread, and the regulatory structures have not necessarily kept pace with the fraudsters.

Machine learning can analyse vast datasets more quickly and accurately than traditional methods. However, it struggles with the non-linear aspects of financial data and in particular textual rather than numeric information. As such, applying advancements in deep learning could bolster machine learning and allow qualitative text found in corporate reports, such as the Management Discussion and Analysis (MD&A) section to be "understood" by fraud-detecting algorithms that might then spot the telltale signs of problematic corporate activity.

Chen's dual-layer approach brings together financial data analysis and sentiment analysis. The use of bidirectional long short-term memory (BiLSTM) networks allows the system to interpret sequences of data, while a parallel network refines the key financial indicators using a convolutional neural network (CNN). Inconsistencies between the sentiment and the financial data can then be revealed. Tests showed a fraud-detection accuracy of 91.35%, with an "Area Under the Curve" of 98.52%. This surpasses traditional fraud-detection methods by a long way, Chen's results suggest.

Chen, W. (2024) 'Financial fraud recognition based on deep learning and textual feature', Int. J. Information and Communication Technology, Vol. 25, No. 12, pp.1–15.
DOI: 10.1504/IJICT.2024.143633

A new method for classifying calligraphy and painting images could be used in the management of cultural heritage, according to research published in the International Journal of Information and Communication Technology.

Nannan Xu OF Suzhou University in Suzhou, China, explains how technology is playing an ever useful role in the preservation and study of artwork and so there is a growing need to find recognition and categorisation tools. The work points out how there is an imbalance in the sample categories that can skew classification models, making it harder to achieve accuracy, and offers a novel solution to this problem. One that could improve accuracy and increase the versatility of image classification for artworks.

Xu introduces a classification method that builds on the AdaBoost algorithm. This machine learning tool works by combining multiple weak classifiers into a strong model and is bolstered by a dynamic training subset construction strategy (DWSCS). According to the research, this approach overcomes the imbalance wherein certain artistic styles are underrepresented. By using sample weights and adjusting how the model is trained on each subset of data, the new method overcomes this bias and so allows a more generalized approach to categorisation where rare artistic styles can be considered.

In cultural heritage, the management and preservation of artworks is critical. This new approach could streamline the cataloguing process for museums and galleries by automating the classification of diverse images. The potential is there for institutions to be able to handle large volumes of calligraphy and paintings efficiently. The same technology might also be useful not only in conservation but in education, offering art historians and students an easier way to analyse and understand the diverse techniques used across different periods and cultures.

Beyond the galleries, the technology might also be used in provenance and authenticity. The system could offer an objective, technology-driven method for verifying the origins of artworks, supporting trust in transactions and authentication processes for art collectors and investors.

Xu, N. (2024) 'Intelligent judgement of calligraphy and painting image categories based on integrated classifier learning', Int. J. Information and Communication Technology, Vol. 25, No. 11, pp.1–20.
DOI: 10.1504/IJICT.2024.143414

A new method for managing urban traffic at multi-intersection networks is discussed in the International Journal of Information and Communication Technology. The research promises improvements in efficiency and adaptability, and by combining technologies could address the long-standing challenges of congestion and unpredictable traffic patterns in dense urban areas.

Renyong Zhang, Shibiao He, and Peng Lu of the Chongqing Institute of Engineering in Chongqing, China, suggest the use of vehicle-to-everything (V2X) technology could allow vehicles and infrastructure to exchange real-time data about road conditions and traffic. This continuous sharing of data would improve the way in which traffic management systems control traffic lights and speed and lane restrictions to smooth the flow of vehicles safely.

The system suggested by the team uses an improved long short-term memory (LSTM) model, a type of artificial intelligence designed for recognizing patterns and making predictions. By using a "sliding time window" update mechanism, the model can learn from real-time data while maintaining historical context. By balancing the two, faster adjustments to traffic flow can be made while reducing the overall computational load on the system and cutting prediction times in half.

The team has carried out simulations and demonstrated that such an approach might reduce average vehicle delays by just under a third and increase road "throughput" by almost 15 percent. The result would be shorter travel times and smoother traffic flow. This should also improve fuel consumption and reduce overall vehicle emissions.

Conventional traffic management systems use historical data or limited real-time inputs, and so cannot respond to actual road conditions at a given time without manual input. Such systems are useful in less complex traffic scenarios, but struggle to handle rapid and unpredictable changes in traffic, particularly in larger, interconnected networks. The newly proposed system addresses these limitations by offering more responsive and precise adjustments.

Zhang, R., He, S. and Lu, P. (2024) 'Multi-intersection traffic flow prediction control based on vehicle-road collaboration V2X and improved LSTM', Int. J. Information and Communication Technology, Vol. 25, No. 11, pp.52–68.
DOI: 10.1504/IJICT.2024.143411