This is a fascinating paper for anyone interested in network analysis or the chronology and cultures of the case study, namely the Late prehistoric burial sites in Dorset, for which the author’s approach allowed a new perspective over an already deeply studied area [1]. This paper's implementation of Similarity Network Fusion (SNF) is noteworthy. This method is typically utilized within genetic research but has yet to be employed in Archaeology. SNF has the potential to benefit Archaeology due to its unique capabilities and approach significantly. 

The author exhibits a deep and thorough understanding of previous investigations concerning material and similarity networks while emphasizing the innovative nature of this particular study. The SNF approach intends to improve a lack of the most used (in Archaeology) similarity coefficient, the Brainerd-Robinson, in certain situations, mainly in heterogenous and noisy datasets containing a small number of samples but a large number of measurements, scale differences, and collection biases, among other things. The SNF technique, demonstrated in the case study, effectively incorporates various similarity networks derived from different datatypes into one network. 

As shown during the Dorset case study, the SNF application has a great application in archaeology, even in already available data, allowing us to go further and bring new visions to the existing interpretations. As stated by the author, SNF shows its potential for other applications and fields in archaeology coping with similar datasets, such as archaeobotany or archaeozoology, and seems to complement different multivariate statistical approaches, such as correspondence or cluster analysis.

This paper has been subject to two excellent revisions, which the author mostly accepted. One of the revisions was more technical, improving the article in the metadata part, data availability and clarification, etc. Although the second revision was more conceptual and gave some excellent technical inputs, it focused more on complementary aspects that will allow the paper to reach a wider audience. I vividly recommend its publication.

References

[1] Geitlinger, T. (2024). Similarity Network Fusion: Understanding Patterns and their Spatial Significance in Archaeological Datasets. Zenodo, 7998239, ver. 3 peer-reviewed and recommended by Peer Community in Archaeology. https://doi.org/10.5281/zenodo.7998239

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