This paper [1] presents an innovative approach to address the prevalent challenge of unbalanced datasets in coin type recognition, shifting the focus from coin class type recognition to coin mint recognition. Despite this shift, the issue of unbalanced data persists. To mitigate this, the authors introduce a method to split larger classes into smaller ones, integrating them into an 'additional class layer'.

Three distinct machine learning (ML) methodologies were employed to identify new possible classes, with one approach utilising unsupervised clustering alongside manual intervention, while the others leverage object detection, and Natural Language Processing (NLP) techniques. However, despite these efforts, overfitting remained a persistent issue, prompting the authors to explore alternative methods such as dataset improvement and Generative Adversarial Networks (GANs).

The paper contributes significantly to the intersection of ML techniques and archaeology, particularly in addressing overfitting challenges. Furthermore, the authors' candid acknowledgment of the limitations of their approaches serves as a valuable resource for researchers encountering similar obstacles.

This study stems from the D4N4 project, aimed at developing a machine learning-based coin recognition model for the extensive "Corpus Nummorum" dataset, comprising over 19,600 coin types and 49,000 coins from various ancient landscapes. Despite encountering challenges with overfitting due to the dataset's imbalance, the authors' exploration of multiple methodologies and transparent documentation of their limitations enriches the academic discourse and provides a foundation for future research in this field.

Reference

[1] Gampe, S. and Tolle, K. (2024). Creating an Additional Class Layer with Machine Learning to counter Overfitting in an Unbalanced Ancient Coin Dataset. Zenodo, 8298077, ver. 4 peer-reviewed and recommended by Peer Community in Archaeology. https://doi.org/10.5281/zenodo.8298077

The paper 'A database of lapidary artifacts in the Caribbean for the Ceramic Age' by Queffelec et al. [1] presents a description of a dataset of nearly 5000 lapidary artefacts from over 100 sites. The data are dominated by beads and pendants, which are mostly made from Diorite, Turquoise, Carnelian, Amethyst, and Serpentine. The raw material data is especially valuable as many of these are not locally available on the island. This holds great potential for exchange network analysis. The data may be especially useful for investigating one of the fundamental questions of this region: whether the Cedrosan and Huecan are separate, little related developments, with different origins, or variants or a single tradition [2].

In addition to metric and technological details about the artefacts, the data include a variety of locational details, including coordinates, distance to coast, and altitude. This enables many opportunities for future spatial analysis and geostatistical modelling to understand human behaviours relating to ornament production, use, and discard. I recommend the authors make a minor revision to Table 1 (spatial coverage of the dataset) to make the column with the citations conform to the same citation style used in the rest of the text.

I warmly commend the authors for making transparency and reproducibility a priority when preparing their manuscript. Their use of the R Markdown format for writing reproducible, dynamic documents [3] is highly impressive. This is an excellent example for others in the international archaeological science community to follow. The paper is especially useful for researchers who are new to R and R Markdown because of the elegant and accessible way the authors document their research here.

[1] Queffelec, A., Fouéré, P. and Caverne, J.-B. 2021. A database of lapidary artifacts in the Caribbean for the Ceramic Age. SocArXiv, 7dq3b, ver. 4 Peer-reviewed and recommended by PCI Archaeology. https://doi.org/10.31235/osf.io/7dq3b

[2] Reed, J. A. and Petersen, J. B. 2001. A comparison of Huecan and Cedrosan Saladoid ceramics at the Trants site, Montserrat. In Proceedings of the XVIIIth International Congress for Caribbean Archaeology (pp. 253-267).

[3] Marwick, B. 2017. Computational Reproducibility in Archaeological Research: Basic Principles and a Case Study of Their Implementation. Journal of Archaeological Method and Theory 24, 424–450. https://doi.org/10.1007/s10816-015-9272-9

“Research perspectives and their influence for typologies” by E. Giannichedda (1) is a contribution to the upcoming volume on the role of typology and type-thinking in current archaeological theory and praxis edited by the recommenders. Taking a decidedly Italian perspective on classificatory practice grounded in what the author dubs the “history of material culture”, Giannichedda offers an inventory of six divergent but overall complementary modes of ordering archaeological material: i) chrono-typological and culture-historical, ii) techno-anthropological, iii) social, iv) socio-economic and v) cognitive. These various lenses broadly align with similarly labeled perspectives on the archaeological record more generally. According to the author, they lend themselves to different ways of identifying and using types in archaeological work. Importantly, Giannichedda reminds us that no ordering practice is a neutral act and typologies should not be devised for their own sake but because we have specific epistemic interests. Even though this view is certainly not shared by everyone involved in the broader debate on the purpose and goal of systematics, classification, typology or archaeological taxonomy (2–4), the paper emphatically defends the long-standing idea that ordering practices are not suitable to elucidate the structure and composition of reality but instead devise tools to answer certain questions or help investigate certain dimensions of complex past realities. This position considers typologies as conceptual prosthetics of knowing, a view that broadly resonates with what is referred to as epistemic instrumentalism in the philosophy of science (5, 6). Types and type-work should accordingly reflect well-defined means-end relationships.

Based on the recognition of archaeology as part of an integrated “history of material culture” rooted in a blend of continental and Anglophone theories, Giannichedda argues that type-work should pay attention to relevant relations between various artefacts in a given historical context that help further historical understanding. Classificatory practice in archaeology – the ordering of artefactual materials according to properties – must thus proceed with the goal of multifaceted “historical reconstruction in mind”. It should serve this reconstruction, and not the other way around. By drawing on the example of a Medieval nunnery in the Piedmont region of northwestern Italy, Giannichedda explores how different goals of classification and typo-praxis (linked to i-v; see above) foreground different aspects, features, and relations of archaeological materials and as such allow to pinpoint and examine different constellations of archaeological objects. He argues that archaeological typo-praxis, for this reason, should almost never concern itself with isolated artefacts but should take into account broader historical assemblages of artefacts. This does not necessarily mean to pay equal attention to all available artefacts and materials, however. To the contrary, in many cases, it is necessary to recognize that some artefacts and some features are more important than others as anchors grouping materials and establishing relations with other objects. An example are so-called ‘barometer objects’ (7) or unique pieces which often have exceptional informational value but can easily be overlooked when only shared features are taken into consideration. As Giannichedda reminds us, considering all objects and properties equally is also a normative decision and does not render ordering less subjective. The archaeological analysis of types should therefore always be complemented by an examination of variants, even if some of these variants are idiosyncratic or even unique. A type, then, may be difficult to define universally.

In total, “Research perspectives and their influence for typologies” emphasizes the need for “elastic” and “flexible” approaches to archaeological types and typologies in order to effectively respond to the manifold research interests cultivated by archaeologists as well as the many and complex past realities they face. Complexity is taken here to indicate that no single research perspective and associated mode of ordering can adequately capture the dimensionality and richness of these past realities and we can therefore only benefit from multiple co-existing ways of grouping and relating archaeological artefacts. Different logics of grouping may simply reveal different aspects of these realities. As such, Giannichedda’s proposal can be read as a formulation of the now classic pluralism thesis (8–11) – that only a plurality of ways of ordering and interrelating artefacts can unlock the full suite of relationships within historical assemblages archaeologists are interested in.

Bibliography

1. Giannichedda, E. (2023). Research perspectives and their influence for typologies, Zenodo, 7322855, ver. 9 peer-reviewed and recommended by Peer Community in Archaeology. https://doi.org/10.5281/zenodo.7322855

2. Dunnell, R. C. (2002). Systematics in Prehistory, Illustrated Edition (The Blackburn Press, 2002).

3. Reynolds, N. and Riede, F. (2019). House of cards: cultural taxonomy and the study of the European Upper Palaeolithic. Antiquity 93, 1350–1358. https://doi.org/10.15184/aqy.2019.49

4. Lyman, R. L. (2021). On the Importance of Systematics to Archaeological Research: the Covariation of Typological Diversity and Morphological Disparity. J Paleo Arch 4, 3. https://doi.org/10.1007/s41982-021-00077-6

5. Van Fraassen, B. C. (2002). The empirical stance (Yale University Press).

6. Stanford, P. K. (2006). Exceeding Our Grasp: Science, History, and the Problem of Unconceived Alternatives (Oxford University Press). https://doi.org/10.1093/0195174089.001.0001

7. Radohs, L. (2023). Urban elite culture: a methodological study of aristocracy and civic elites in sea-trading towns of the southwestern Baltic (12th-14th c.) (Böhlau).

8. Kellert, S. H., Longino, H. E. and Waters, C. K. (2006). Scientific pluralism (University of Minnesota Press).

9. Cat, J. (2012). Essay Review: Scientific Pluralism. Philosophy of Science 79, 317–325. https://doi.org/10.1086/664747

10. Chang, H. (2012). Is Water H2O?: Evidence, Realism and Pluralism (Springer Netherlands). https://doi.org/10.1007/978-94-007-3932-1

11. Wylie, A. (2015). “A plurality of pluralisms: Collaborative practice in archaeology” in Objectivity in Science, (Springer), pp. 189–210. https://doi.org/10.1007/978-3-319-14349-1_10

This paper (2023) describes The Living Hill project dedicated to the archaeological park and fortress of Poggio Imperiale in Poggibonsi, Italy. The project is a collaboration between the Poggibonsi excavation and Entertainment Games App, Ltd. From the start, the project focused on the question of the intended audience rather than on the used technology. It was therefore planned to involve the audience in the creation of the game itself, which was not possible after all due to the covid pandemic. Nevertheless, the game aimed towards a visit experience as close as possible to reality to offer an educational tool through the video game, as it offers more periods than the medieval period showcased in the archaeological park itself.

The game mechanics differ from a walking simulator, or a virtual tour and the player is tasked with returning three lost objects in the virtual game. While the medieval level was based on a 3D scan of the archaeological park, the other two levels were reconstructed based on archaeological material. Currently, only a PC version is working, but the team works on a mobile version as well and teased the possibility that the source code will be made available open source. Lastly, the team also evaluated the game and its perception through surveys, interviews, and focus groups. Although the surveys were only based on 21 persons, the results came back positive overall.

The paper is well-written and follows a consistent structure. The authors clearly define the goals and setting of the project and how they developed and evaluated the game. Although it has be criticized that the game is not playable yet and the size of the questionnaire is too low, the authors clearly replied to the reviews and clarified the situation on both matters. They also attended to nearly all of the reviewers demands and answered them concisely in their response. In my personal opinion, I can fully recommend this paper for publication.

For future works, it is recommended that the authors enlarge their audience for the quesstionaire in order to get more representative results. It it also recommended to make the game available as soon as possible also outside of the archaeological park. I would also like to thank the reviewers for their concise and constructive criticism to this paper as well as for their time.

References

Mariotti, Samanta. (2023) Transforming the Archaeological Record Into a Digital Playground: a Methodological Analysis of The Living Hill Project, Zenodo, 8302563, ver. 5 peer-reviewed and recommended by Peer Community in Archaeology https://doi.org/10.5281/zenodo.8302563

This recommended paper on the IUENNA project (Hagmann and Reiner 2023) is not a paper in the traditional sense, but it is a reworked version of a project proposal. It is refreshing to read about a project that has just started and see what the aims of the project are. This ties in with several open science ideas and standards (e.g. Brinkman et al. 2023). I am looking forward to see in a few years how the authors managed to reach the aims and goals of the project.

The IUENNA project deals with the legacy data and old excavations on the Hemmaberg and in the Jauntal. Archaeological research in this small, but important region, has taken place for more than a century, revealing material from over 2000 years of human history. The Hemmaberg is well known for its late antique and early medieval structures, such as roads, villas and the various churches. The wider Jauntal reveals archaeological finds and features dating from the Iron Age to the recent past. The authors of the paper show the need to make sure that the documentation and data of these past archaeological studies and projects will be accessible in the future, or in their own words: "Acute action is needed to systematically transition these datasets from physical filing cabinets to a sustainable, networked virtual environment for long-term use" (Hagmann and Reiner 2023: 5).

The papers clearly shows how this initiative fits within larger developments in both Digital Archaeology and the Digital Humanities. In addition, the project is well grounded within Austrian archaeology. While the project ties in with various international standards and initiatives, such as Ariadne (https://ariadne-infrastructure.eu/) and FAIR-data standards (Wilkinson et al. 2016, 2019), it would benefit from the long experience institutes as the ADS (https://archaeologydataservice.ac.uk/) and DANS (see Data Station Archaeology: https://dans.knaw.nl/en/data-stations/archaeology/) have on the storage of archaeological data. I would also like to suggest to have a look at the Dutch SIKB0102 standard (https://www.sikb.nl/datastandaarden/richtlijnen/sikb0102) for the exchange of archaeological data. The documentation is all in Dutch, but we wrote an English paper a few years back that explains the various concepts (Boasson and Visser 2017). However, these are a minor details or improvements compared to what the authors show in their project proposal. The integration of many standards in the project and the use of open software in a well-defined process is recommendable.

The IUENNA project is an ambitious project, which will hopefully lead to better insights, guidelines and workflows on dealing with legacy data or documentation. These lessons will hopefully benefit archaeology as a discipline. This is important, because various (European) countries are dealing with similar problem, since many excavations of the past have never been properly published, digitalized or deposited. In the Netherlands, for example, various projects dealt with publication of legacy excavations in the Odyssee-project (https://www.nwo.nl/onderzoeksprogrammas/odyssee). This has led to the publication of various books and datasets (24) (https://easy.dans.knaw.nl/ui/datasets/id/easy-dataset:34359), but there are still many datasets (8) missing from the various projects. In addition, each project followed their own standards in creating digital data, while IUENNA will make an effort to standardize this. There are still more than 1000 Dutch legacy excavations still waiting to be published and made into a modern dataset (Kleijne 2010) and this is probably the case in many other countries. I sincerely hope that a successful end of IUENNA will be an inspiration for other regions and countries for future safekeeping of legacy data.

References

Boasson, W and Visser, RM. 2017 SIKB0102: Synchronizing Excavation Data for Preservation and Re-Use. Studies in Digital Heritage 1(2): 206–224. https://doi.org/10.14434/sdh.v1i2.23262

Brinkman, L, Dijk, E, Jonge, H de, Loorbach, N and Rutten, D. 2023 Open Science: A Practical Guide for Early-Career Researchers https://doi.org/10.5281/zenodo.7716153

Hagmann, D and Reiner, F. 2023 IUENNA – openIng the soUthErn jauNtal as a micro-regioN for future Archaeology: A ‘para-description’. https://doi.org/10.31219/osf.io/5vwg8

Kleijne, JP. 2010. Odysee in de breedte. Verslag van het NWO Odyssee programma, kortlopend onderzoek: ‘Odyssee, een oplossing in de breedte: de 1000 onuitgewerkte sites, die tot een substantiële kennisvermeerdering kunnen leiden, digitaal beschikbaar!’ ‐ ODYK‐09‐13. Den Haag: E‐depot Nederlandse Archeologie (EDNA). https://doi.org/10.17026/dans-z25-g4jw

Wilkinson, MD, Dumontier, M, Aalbersberg, IjJ, Appleton, G, Axton, M, Baak, A, Blomberg, N, Boiten, J-W, da Silva Santos, LB, Bourne, PE, Bouwman, J, Brookes, AJ, Clark, T, Crosas, M, Dillo, I, Dumon, O, Edmunds, S, Evelo, CT, Finkers, R, Gonzalez-Beltran, A, Gray, AJG, Groth, P, Goble, C, Grethe, JS, Heringa, J, ’t Hoen, PAC, Hooft, R, Kuhn, T, Kok, R, Kok, J, Lusher, SJ, Martone, ME, Mons, A, Packer, AL, Persson, B, Rocca-Serra, P, Roos, M, van Schaik, R, Sansone, S-A, Schultes, E, Sengstag, T, Slater, T, Strawn, G, Swertz, MA, Thompson, M, van der Lei, J, van Mulligen, E, Velterop, J, Waagmeester, A, Wittenburg, P, Wolstencroft, K, Zhao, J and Mons, B. 2016 The FAIR Guiding Principles for scientific data management and stewardship. Scientific Data 3(1): 160018. https://doi.org/10.1038/sdata.2016.18

Wilkinson, MD, Dumontier, M, Jan Aalbersberg, I, Appleton, G, Axton, M, Baak, A, Blomberg, N, Boiten, J-W, da Silva Santos, LB, Bourne, PE, Bouwman, J, Brookes, AJ, Clark, T, Crosas, M, Dillo, I, Dumon, O, Edmunds, S, Evelo, CT, Finkers, R, Gonzalez-Beltran, A, Gray, AJG, Groth, P, Goble, C, Grethe, JS, Heringa, J, Hoen, PAC ’t, Hooft, R, Kuhn, T, Kok, R, Kok, J, Lusher, SJ, Martone, ME, Mons, A, Packer, AL, Persson, B, Rocca-Serra, P, Roos, M, van Schaik, R, Sansone, S-A, Schultes, E, Sengstag, T, Slater, T, Strawn, G, Swertz, MA, Thompson, M, van der Lei, J, van Mulligen, E, Jan Velterop, Waagmeester, A, Wittenburg, P, Wolstencroft, K, Zhao, J and Mons, B. 2019 Addendum: The FAIR Guiding Principles for scientific data management and stewardship. Scientific Data 6(1): 6. https://doi.org/10.1038/s41597-019-0009-6