Dura-Europos Stories: Explore the Excavations and Artifacts

Review:

1. Originality: The paper introduces a unique and innovative approach to exploring archaeological artifacts through the Dura-Europos stories multimedia application, leveraging the Wikidata data model and knowledge graphs.

2. Significance of the Results: The development of Dura-Europos stories offers a valuable platform for engaging with archaeological artifacts and facilitates the combination of project-specific data with other relevant information on the web.

3. Quality of Writing: The paper is well-written, effectively conveying the purpose, methodology, and outcomes of the Dura-Europos stories project. The language is clear, and the structure enables easy comprehension. Minor revisions are suggested for formatting footnotes 16 and 17, as they exceed the margin.

4. Suggestions: The use of Wikidata as a structured data format for this project is commendable. However, it is suggested to also consider incorporating parameteric knowledge stored in large language models such as ChatGPT and LLaMA. Designing task-specific prompts to extract knowledge from these models can generate natural language outputs, enabling more engaging descriptions and related information about the artifacts. By modifying the prompt, the style of generation can be controlled, enhancing the overall user experience. This would open up additional possibilities, such as enabling users to have interactive conversations with the system about the artifacts and allowing the system to provide answers to user questions, thereby enriching the user experience.

Overall, the paper is recommended for acceptance with minor revision.

This method paper introduces a new multimedia application for cataloguing artefacts and places related to Dura-Europos archaeological site. The application combines data from multiple sources and uses semantic technologies (Wikidata) to facilitate historical explorations. The authors claim that the advantage of using Wikidata are multiple:
- it has a large number of contributors compared to project-specific databases
- the data is continuously maintained
- it supports multiple languages
- Wikidata's graph of external identifiers can be leveraged to access more information sources

The authors mentioned a number of previous approaches in digital humanities that either use semantic technologies and/or Wikidata, as well as 3D displays.

While I am not an expert in either semantic web or digital humanities I considered the approach interesting and the results appear compelling to me. The paper is well-structured and generally well-written with an abundance of details, although some parts of the paper would need, in my opinion, further details and clarifications.

The literature review section (Section 1) references similar approaches while omitting to address the pros and cons of each of them. I feel the paper would benefit from a more in-depth comparison and highlight of what the proposed approach adds to the existing literature beyond the obvious application to the Dura-Europos archaeological site.

I am also unsure why so many detail was granted to the work performed by the International Digital Dura-Europos Archive (IDEA) team (Section 5). While it is important to understand how the data used by the application came about, I feel that the space could have been used more efficiently to expand on the core application originality and functionalities.

Overall, the paper describes an interesting application of semantic web data in the field of digital humanities. Although the present contribution will undoubtedly be beneficial to Dura-Europos archaeologists, the paper is missing more generalisation for scenarios where other researchers would embark on a similar project for a different archaeological site. Such generalisation would include pitfalls to avoid, things to look out for and takeaways from the undertaking.