Search results

This paper provides practical advice for archaeologists and heritage specialists wishing to use ML approaches to identify archaeological features in high-resolution satellite imagery (or other remotely sensed data sources). We seek to balance the disproportionately optimistic literature related to the application of ML to archaeological prospection through a discussion of limitations, challenges and other difficulties. We further seek to raise awareness among researchers of the time, effort, expertise and resources necessary to implement ML successfully, so that they can make an informed choice between ML and manual inspection approaches.

Automated object detection has been the holy grail of archaeological remote sensing for the last two decades. Machine learning (ML) models have proven able to detect uniform features across a consistent background, but more variegated imagery remains a challenge. We set out to detect burial mounds in satellite imagery from a diverse landscape in Central Bulgaria using a pre-trained Convolutional Neural Network (CNN) plus additional but low-touch training to improve performance. Training was accomplished using MOUND/NOT MOUND cutouts, and the model assessed arbitrary tiles of the same size from the image. Results were assessed using field data.

Validation of results against field data showed that self-reported success rates were misleadingly high, and that the model was misidentifying most features. Setting an identification threshold at 60% probability, and noting that we used an approach where the CNN assessed tiles of a fixed size, tile-based false negative rates were 95–96%, false positive rates were 87–95% of tagged tiles, while true positives were only 5–13%. Counterintuitively, the model provided with training data selected for highly visible mounds (rather than all mounds) performed worse. Development of the model, meanwhile, required approximately 135 person-hours of work.

Our attempt to deploy a pre-trained CNN demonstrates the limitations of this approach when it is used to detect varied features of different sizes within a heterogeneous landscape that contains confounding natural and modern features, such as roads, forests and field boundaries. The model has detected incidental features rather than the mounds themselves, making external validation with field data an essential part of CNN workflows. Correcting the model would require refining the training data as well as adopting different approaches to model choice and execution, raising the computational requirements beyond the level of most cultural heritage practitioners.

Improving the pre-trained model’s performance would require considerable time and resources, on top of the time already invested. The degree of manual intervention required – particularly around the subsetting and annotation of training data – is so significant that it raises the question of whether it would be more efficient to identify all of the mounds manually, either through brute-force inspection by experts or by crowdsourcing the analysis to trained – or even untrained – volunteers. Researchers and heritage specialists seeking efficient methods for extracting features from remotely sensed data should weigh the costs and benefits of ML versus manual approaches carefully.

Our literature review indicates that use of artificial intelligence (AI) and ML approaches to archaeological prospection have grown exponentially in the past decade, approaching adoption levels associated with “crossing the chasm” from innovators and early adopters to the majority of researchers. The literature itself, however, is overwhelmingly positive, reflecting some combination of publication bias and a rhetoric of unconditional success. This paper presents the failure of a good-faith attempt to utilise these approaches as a counterbalance and cautionary tale to potential adopters of the technology. Early-majority adopters may find ML difficult to implement effectively in real-life scenarios.

Unlike many high-profile reports from well-funded projects, our paper represents a serious but modestly resourced attempt to apply an ML approach to archaeological remote sensing, using techniques like transfer learning that are promoted as solutions to time and cost problems associated with, e.g. annotating and manipulating training data. While the majority of articles uncritically promote ML, or only discuss how challenges were overcome, our paper investigates how – despite reasonable self-reported scores – the model failed to locate the target features when compared to field data. We also present time, expertise and resourcing requirements, a rarity in ML-for-archaeology publications.

Data storytelling courses position students as agents in creating stories interpreted from data about a social problem or social justice issue. The purpose of this study is to explore two research questions: What themes characterized students’ iterative development of data story topics? Looking back at six years of iterative feedback, what categories of data literacy pedagogy did instructors engage for these themes?.

This project examines six years of data storytelling final projects using thematic analysis and three years of instructor feedback. Ten themes in final projects align with patterns in feedback. Reflections on pedagogical approaches to students’ topic development suggest extending data literacy pedagogy categories – formal, personal and folk (Pangrazio and Sefton-Green, 2020).

Data storytelling can develop students’ abilities to move from being consumers to creators of data and interpretations. The specific topic of personal data exposure or risk has presented some challenges for data literacy instruction (Bowler et al., 2017). What “personal” means in terms of data should be defined more broadly. Extending the data literacy pedagogy categories of formal, personal and folk (Pangrazio and Sefton-Green, 2020) could more effectively center social justice in data literacy instruction.

Implications for practice include positioning students as producers of data interpretation, such as role-playing data analysis or decision-making scenarios.

Data storytelling has the potential to address current challenges in data literacy pedagogy and in teaching critical data literacy.

Course descriptions provide a template for future data literacy pedagogy involving data storytelling, and findings suggest implications for expanding definitions and applications of personal and folk data literacies.

This study aims to develop an extended social attachment model for expatriates, integrating a multiple stakeholder perspective, to understand evacuation decisions during disasters.

Through interviews with 12 Tokyo-based expatriates who experienced the 2011 Tohoku earthquake, tsunami and nuclear disasters, this study collects the lived experiences of a diverse set of expatriates. This data is analyzed abductively to map relevant evacuation factors and to propose a reaction typology.

While the 2011 Tohoku disasters caused regional destruction and fears of nuclear fallout, Tokyo remained largely unscathed. Still, many expatriates based in Tokyo chose to leave the country. Evacuation decisions were shaped by an interplay of threat assessment, location of attachment figures and cross-cultural adjustment. The study also discusses the influence of expatriate types.

Disaster planning is often overlooked or designed primarily with host country nationals in mind. Expatriates often lack the disaster experience and readiness of host country nationals in disaster-prone regions in Asia and beyond, and thus might need special attention when disaster strikes. This study provides advice for how to do so.

By unpacking the under-researched and complex phenomenon of expatriate reactions to disasters, this study contributes to the fields of international human resource and disaster management. Specifically, seven proposition on casual links leading to expatriate evacuation are suggested, paving the way for future research.

Amidst the complicated nature of the UAE’s facilities management (FM) industry, the need to recalibrate the existing performance measurement (PM) system measures and criteria has been resonating to ensure their ability to capture the FM industry trends and dynamics, thus enhancing organizational excellence. Therefore, this research aimed to propose a specific PM tool to the country’s FM industry to accurately assess performance and establish strategic enhancements.

The study reviewed literature on the available PM systems to gather the available measures, which were presented to a focus group of seven participants, who were purposively selected based on their expertise in FM and PM implementation in the UAE to adjust them and add ones relevant to the UAE’s FM industry.

The focus group conducted various changes, from retaining certain measures and criteria, renaming them to simplify or make them more representative of the industry, ranking them based on their importance to limit their numbers, to finally categorizing them as enablers or results. Consequently, the final proposed tool was composed of nine dimensions with 51 measures as performance enablers and three dimensions with 11 measures as performance results. Seven measures were added by the experts, who highlighted their increasing popularity in the UAE’s FM industry.

Through addressing the critical void in literature, this paper develops a specific PM tool aligning with the intricacy of the UAE’s FM industry, thus providing proactive contribution to the industry’s effective and sustainable growth.

This study aims to examine the historical development, present state and potential future directions of the integration between building information modeling (BIM) and life cycle assessment (LCA) in the field of construction. Additionally, this paper identifies current problems while offering insight into worldwide BIM research trends.

This study uses text mining on unstructured abstracts, a novel approach not previously documented in BIM research. By conducting a comprehensive systematic assessment of academic literature, this work uses advanced bibliometric approaches to examine the developmental trajectory of the integration of BIM and LCA. The research incorporates co-citation and keyword co-occurrence mapping, providing a complex visual depiction of the interconnectedness of information across different periods.

The results of this analysis reveal the historical development of the integration of BIM and LCA, including its roots and the initial research that established the foundation for further investigations. The aforementioned seminal works signify the inception of the discipline, serving as a source of inspiration for current scholarly investigations. Currently, there is a complex network of interdisciplinary cooperation that can be observed, combining knowledge and perspectives from the fields of design, engineering, construction and sustainability.

This research contributes novelty to the scholarly discourse by offering a holistic and up-to-date panorama of the dynamic BIM and LCA research landscape. It identifies emerging trends, influential contributors and uncharted territories, thus providing a foundation for scholars to contribute meaningfully to the advancement of knowledge in sustainable construction practices.

Considering food waste as a global problem resulting from the wastage of valuable resources that could fulfil the requirements of malnourished people, the current research focusses on understanding consumerism’s impact on this phenomenon. Additionally, the circular economy (CE) approach can be critical in reducing food waste and promoting sustainability.

A systematic literature review was conducted using bibliometrics and network analysis. The study reviewed 326 articles within 10 years, from 2013 to 2023.

The findings reveal four prominent factors – behavioural, environmental, socioeconomic and technological – in managing food waste (FW). Reducing FW at a holistic level can benefit individuals and the environment in several ways.

Consumers are encouraged to be more responsible for their food consumption by reducing food waste, as it affects societies and businesses both economically and environmentally. This can help promote a responsible consumption culture that values quality over quantity and encourages people to make more informed choices about what they eat and how they dispose of it post-consumption. All stakeholders, including firms, the government and consumers, must examine the motives behind inculcating pro-environmental behaviour.

Addressing consumerism and the ability to decrease FW behaviour are complex issues that require a multidimensional approach. This study seeks to fill the gap in understanding consumerism and the capacity to reduce FW using the CE approach and understand the research gaps and future research trends.