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.

This paper shares findings from a multi-year literacy professional development partnership between a school district and university (2014–2019). We share this case of a Literacy Cohort initiative as an example of cross-institutional professional development situated within several of NAPDS’ nine essentials, including professional learning and leading, boundary-spanning roles and reflection and innovation (NAPDS, 2021).

We asked, “In what ways did the Cohort initiative create conditions for community and collaboration in the service of meaningful literacy reforms?” Drawing on social design methodology (Gutiérrez & Vossoughi, 2010), we sought to generate and examine the educational change associated with this multi-year initiative. Our data set included programmatic data, interviews (N = 30) and artifacts of literacy teaching, learning and leading.

Our findings reflect the emphasis areas that are important to educators in the partnership: diversity by design, building relationships through collaboration and rooting literacy reforms in teacher leadership. Our discussion explores threads of reciprocity, simultaneous renewal and boundary-spanning leadership and their role in sustaining partnerships over time.

This paper contributes to our understanding of building and sustaining a cohort model of multi-year professional development through the voices, perspectives and experiences of teachers, faculty and district administrators.

The purpose of the study is to understand the relationship between family-driven innovation and the incorporation of corporate sustainability in German family firms.

The study conducted 26 interviews with 22 German family firms. Thematic analysis was undertaken on the collected data resulting in five major themes.

The study identified five main themes of corporate sustainability-oriented innovation in family firms, which include measuring corporate sustainability performances, building corporate sustainability-oriented infrastructure, stabilizing/optimizing operations, enhancing operational flexibility/independence and knowledge management and development. The study also provides an activity-based guide for family firms to use innovation to achieve corporate sustainability goals and present the findings’ implications for policymakers.

The present study is the first study to empirically investigate the relationship between family-driven innovation and the incorporation of corporate sustainability at each of the corporate sustainability maturity levels.

The purpose of this paper is to analyze the incorporation, prioritization and depth of equity, diversity and inclusion (EDI) initiatives in tourism industry restart policies of Canadian provinces and territories. This study investigates how the detailing of EDI in policies determine the priority in emancipating tourism workers from the inequities exacerbated during the pandemic. Such investigation enables a better understanding of the complexities, tendencies and rationale of involving EDI in the tourism industry’s recovery.

The research investigated the presence and prioritization of equity, diversity, and inclusion using systematic text analytics of 38 publicly available restart plans and statements from 52 government and non-government agencies. Using web-based software Voyant Tools to assist in text analytics, a hybrid deductive-inductive coding approach was conducted.

Key outcomes from the analysis revealed scarce to no full and dedicated content on EDI as a holistic initiative necessary for tourism industry relaunch. This lack of EDI content was a result of the greater impetus to prioritize economic generation and limited data due to practical and ideological issues. Results also suggested the tokenizing of EDI in some policies.

Difficulties in data used for research include the lack and availability of restart policies specifically for tourism; most policies were generalized and referred to economic recovery as a whole. Studies of tourism-specific EDI issues were also limited.

The research is revelatory for investigating EDI prioritizations in restart policies even among well-developed and worker-diverse tourism industries such as in Canada, where inequities and injustices to women, Black, Indigenous, gender-diverse, and newcomer tourism workers among others have been withstanding.

Despite its devastating nature, the COVID-19 pandemic has also catalyzed a substantial surge in the adoption and integration of technological tools within economies, exerting a profound influence on the dissemination of information among participants in stock markets. Consequently, this present study delves into the ramifications of post-pandemic dynamics on stock market behavior. It also examines the relationship between investors' sentiments, underlying behavioral drivers and their collective impact on global stock markets.

Drawing upon data spanning from 2012 to 2023 and encompassing major world indices classified by Morgan Stanley Capital International’s (MSCI) market and regional taxonomy, this study employs a threshold regression model. This model effectively distinguishes the thresholds within these influential factors. To evaluate the statistical significance of variances across these thresholds, a Wald coefficient analysis was applied.

The empirical results highlighted the substantive role that investors' sentiments and behavioral determinants play in shaping the predictability of returns on a global scale. However, their influence on developed economies and the continents of America appears comparatively lower compared with the Asia–Pacific markets. Similarly, the regions characterized by a more pronounced influence of behavioral factors seem to reduce their reliance on these factors in the post-pandemic landscape and vice versa. Interestingly, the post COVID-19 technological advancements also appear to exert a lesser impact on developed nations.

This study pioneers the investigation of these contextual dissimilarities, thereby charting new avenues for subsequent research studies. These insights shed valuable light on the contextualized nexus between technology, societal dynamics, behavioral biases and their collective impact on stock markets. Furthermore, the study's revelations offer a unique vantage point for addressing market inefficiencies by pinpointing the pivotal factors driving such behavioral patterns.