FindSampo Platform for Reporting and Studying Archaeological Finds Using Citizen Science

Abstract

This paper introduces the FindSampo Reporter and Portal platforms for reporting and studying archaeological nds on the Semantic Web, respectively. FindSampo brings together members of the public, scientists, cultural heritage managers, and archaeologists utilising citizen science mediated by Linked Open Data and emerging Web development technologies. Our focus is on reporting technical results on designing the user interfaces and on evaluating the Reporter part in a eld test.5 The popularity of metal detecting has grown rapidly in recent years. Many European countries have therefore started to develop digital reporting services to collect, analyse, and study archaeological data: 1) Portable Antiquities Scheme (PAS)6 records7 archaeological discoveries found by members of the public in England and Wales since 1997 [1]; 2) Digital Metal Finds (DIME)8 is an online platform for reporting metal detecting nds in Denmark [10]; 3) Portable Antiquities of the Netherlands (PAN)9 is an online portal in use in the Netherlands [9]; 4) Metal-Detected Artefacts (MEDEA)10 is an online portal developed in Flanders for metal detectors [2, 10]; 5) ILPPARI11 is a portal of the Finnish Heritage Agency (FHA) for reporting archaeological nds by citizens [11].

Keywords

• Citizen science
• Semantic Web
• Archaeological nds

-

FindSampo is a new research prototype of the SuALT project12 aiming to study and improve the reporting process and analysis of archaeological nds based on collaboration of the public, academic researchers, archaeologists, and the FHA [ 4, 11, 8 ]. The system includes two interlinked applications: 1) FindSampo Reporter, a mobile reporting system for nds, and 2) FindSampo Portal for analysing and studying the nds collection on the Web. The intent is to use citizen science to enable citizens to participate in archaeological research and improve themselves in di erent stages of reporting, and to expedite processing of reports. Such involvement allows participants to improve themselves and also learn more about archaeology. In this way, archaeological data becomes more quickly and comprehensively available and accessible for research purposes and Digital Humanities (DH) [ 3 ].

This paper presents the design principles of the FindSampo Reporter and Portal user interfaces and results of a rst eld test for evaluating the Reporter part with metal detectorists. In contrast to the aforementioned related works, FindSampo is based on Linked Open Data. FindSampo Portal is yet another member in the "Sampo" series13 of Linked Open Data services and semantic portals [ 7 ], based on a national Semantic Web infrastructure [ 6 ]. 2

User Interface Design: FindSampo Reporter and Portal

A user-centred design approach was adopted right from the beginning in the design process of FindSampo Reporter. Requirements were elicited at an early stage using online surveys and interviews, and they have been evaluated continuously through showing and reviewing mock-up interface designs with end users and nally in eld testings [ 8 ].

Another key decision was to enforce a mobile- rst strategy, which optimises the design rstly for mobile devices and afterwards for others. Adopting a mobilerst approach is critical because especially the archaeological nd reporting functionalities are intended to be used in the eld. On the other hand, it is more natural to carry out tasks related to data analysis on a desktop with a large screen, so it is important that the user interface adapts to larger screens also. Figure 1 shows how the user interface adapts to di erent screen sizes in order to reduce the user's cognitive load.

Additionally, in order to maximise the e ciency of the nd reporting process from the citizen's point of view, the need to enter data is minimised. The reporting process consists of short and simple questions that express only one idea and are easy to answer. The process is split into a series of self-contained steps to make the report ow understandable. FindSampo Reporter provides help during the reporting process with the aim of making nd reporting as easy as possible. Such a process improves nders' knowledge of archaeology and would also reduce the workload of FHA's o cers as the quality of the reports improves. Figure 2 shows some example steps of the reporting process above-mentioned: if 12 SuALT project: https://blogs.helsinki. /sualt-project 13 For a list of Sampo portals, see https://seco.cs.aalto. /applications/sampo/. the user is on the nd spot, the coordinate information as well as time can be read automatically from the mobile phone (image on the left); the nd spot can be seen and edited also later on (image in the middle); the depth of the nd can be set with a slider (image on the right).

The idea of minimising the reporting work of the metal detectorists was deemed important in order to foster reporting of nds in the rst place, since according to our interviews, metal detectorists seem to like reporting tools that would record less information. However, from the archaeologists/heritage managers' viewpoint, it is necessary to collect all relevant data about the object and its nd context. So, we are facing here a somewhat challenging situation to develop a platform for all user groups' needs. Our solution proposal in FindSampo Reporter is that additional information can be collected from the detectorists, if needed, after they have provided the basic information of the nd, and a connection between the nder and FHA has been established.

In FindSampo we plan that all validated archaeological objects found in Finland would be openly available to the public with the exact nd spots. This would help metal detectorists to nd promising areas for exploration. Furthermore, opening the nd data in detail, would be useful for Digital Humanities research. For this purpose, the FindSampo Reporter prototype already includes a semantic portal for searching and viewing the nds on maps, with additional linked information from other relevant GIS services of FHA. For example, it is possible to see on maps not only the individual nds but also protected archaeological sites where metal detecting is forbidden by law.

However, disclosing the exact nd spot data is a challenging and critical decision for openness still to be con rmed, as many archaeologists argue that there is the danger that disclosing the exact nd spots would lead to looting. In many countries and systems, such as DIME in Denmark, location data is fuzzi ed before publishing. Furthermore, according to our interviews, the detectorists usually would not like to disclose their nd spots immediately but want to investigate them rst by themselves. The compromise solution we are aiming at the moment is to disclose the exact nd spots only after a period of time, say one year. If the nd spot is deemed important, it is then also possible to add it in the list of protected areas.

The data of validated nds will be accessible via the FindSampo Portal based on the Sampo-UI framework14 and a SPARQL15 endpoint at the Linked Data Finland platform16. The initial version of FindSampo's public Linked Open Data database contains approximately 3600 archaeological nds, and more nds will be added as the FHA delivers new validated data.

The validated nds can be studied through di erent views in the FindSampo Portal, based on the Sampo model [ 7 ]. This model includes three components: 1) A \business model" for harmonizing, aggregating, and publishing heterogeneous, distributed contents based on a shared ontology infrastructure. 2) An 14 Sampo-UI homepage: https://seco.cs.aalto. /tools/sampo-ui/ 15 SPARQL: https://www.w3.org/TR/sparql11-query. 16 Linked Data Finland Portal: http://www.ldf. approach to interface design, where the data can be re-used and accessed independently from multiple application perspectives, while the data resides in a single SPARQL endpoint. 3) A two-step model for accessing and analyzing the data where the focus of interest is rst ltered out using faceted semantic search, and then visualized or analyzed by ready-to-use DH tools of the portal.

An important part of the FindSampo Portal is the underlying ontology infrastructure, a basis of the Sampo model. As a starting point, the MAO/TAO ontology for Museum Domain and Applied Arts of the Finnish ontology infrastructure FinnONTO [ 5 ] available today at the Finto. service17 is used. This ontology is being enriched with new concepts extracted from the nds databases and is being aligned with international ontologies, such as AAT18 and PeriodO19, as part of the ARIADNEplus project20.

In FindSampo Portal, the nds can be analysed and interpreted using faceted search which enables users to analyse, lter, and organise data by applying multiple criteria at the same time. Thus, users can browse a large amount of data based on the speci c search criteria such as object type, material, period, municipality, and province. Furthermore, users can view archaeologically signi cant sites with tutorials to gain knowledge about archaeology and also the rules of law in their pastime. Figure 3 shows the di erent views for visualising archaeological data with faceted search in FindSampo: the Clustered Map view (on the left) is used for providing the user with an aggregated view of ltered nds on the map; the HeatMap view visualizes the distribution of the ltered nds in colors; the Table view lists the nds in a traditional way; the Statistics view illustrates statistical distributions of the nds along di erent facet dimensions, here based on the material of the selected nds. 3

Implementation

In the design of FindSampo, a variety of Web development technologies were analysed and as a result, the most appropriate ones selected for implementing it. The technologies were chosen on the basis of technical e ciency and user needs.

FindSampo Reporter utilises the Semantic Web and emerging Web development technologies to provide a platform for reporting and studying archaeological nds. It is a single-page application and therefore, it does not need page reloading when navigating around the application. Such a structure enables users to use it more e ciently in the places where the Internet connection is slow.

FindSampo uses a set of modern JavaScript libraries such as React21, Redux22, Material UI23, and Sass24 to build the client. The server is implemented 17 http:// nto. /maotao/en/?clang= 18 https://www.getty.edu/research/tools/vocabularies/aat/ 19 https://perio.do/en/ 20 https://ariadne-infrastructure.eu/ 21 React: https://reactjs.org 22 Redux: https://redux.js.org 23 Material UI: https://material-ui.com 24 Sass: https://sass-lang.com using NodeJS25 and ExpressJS26 to enable a lightweight interaction with external services. Semantic Web technologies, such as ontologies, are used for harmonizing data within Finnish archaeological data sources and establishing interoperability with international archaeological resources. 4

Evaluation

To gain in-depth understanding of FindSampo Reporter from a user-centric perspective, a eld testing was performed with a small number of metal detectorists [4, Ch. 7]. After reporting the nds through FindSampo Reporter, the detectorists lled out a user experience survey. As a result, the overall impression of the prototype was very good, and all metal detectorists very likely would recommend it to friends or colleagues in the future. Furthermore, the detectorists were fairly well satis ed with the reporting system. All participants found FindSampo more e cient, attractive, and easier to use than the FHA's current reporting service ILPPARI. The testers in particular liked the mobile-friendly user interface, e cient and e ective reporting process, automatic location identi cation, and search page for validated archaeological nds. 5

Discussion and Future Work

FindSampo takes the current state of archaeological nd databases a step further by providing a mobile platform that combines the advantages of the existing ones and also utilises citizen science, Semantic Web, and emerging Web development technologies. The evaluation of the prototype and user experience surveys reveal that the platform improves signi cantly archaeological data collection, analysis, and interpretation processes. Furthermore, it provides further research opportunities by visualising and linking archaeological data as well as improving its availability and accessibility.

In the future, the members of the public can hopefully participate in all stages of research and report processing using FindSampo and ILPPARI, the legacy system of FHA for nd report management. FindSampo aims at building a community in which everybody gains bene ts, including volunteers, research communities, heritage managers, and also the whole society using citizen science. Furthermore, FindSampo is going to generate data models that are semantically interoperable with existing cultural heritage databases, other European archaeological resources in order to provide research opportunities by visualising archaeological data as well as improving its availability and accessibility. The data will be part of the Pan-European ARIADNEplus infrastructure27.

Acknowledgements. Our work was funded by the Academy of Finland at the University of Helsinki, Aalto University, and the Finnish Heritage Agency. Thanks to CSC { IT Center for Science, Finland, for computational resources. 25 NodeJS: https://nodejs.org 26 ExpressJS: https://expressjs.com 27 https://ariadne-infrastructure.eu

FindSampo: Studying Archaeological Finds Using Citizen Science

Appendix: Figures

Clustered Map Faceted Search on desktop and mobile screens

References

1. Daubney , A. , Nicholas , L.E. : Detecting Heritage Crime(s): What We Know about Illicit Metal Detecting in England and Wales . International Journal of Cultural Property 26 , 139 {165 (May 2019 ). https://doi.org/10.1017/S0940739119000158

2. Deckers , P. , Bleumers , L. , Ruelens , S. , Lemmens , B. , Vanderperren , N. , Marchal , C. , Pierson , J. , Tys , D. : MEDEA: Crowd-Sourcing the Recording of Metal-Detected Artefacts in Flanders (Belgium) . Open Archaeology 2 ( 1 ), 264 {277 ( June 2016 ). https://doi.org/10.1515/opar-2016 -0019

3. Gardiner , E. , Musto , R.G. : The Digital Humanities: A Primer for Students and Scholars . Cambridge University Press, New York, NY, USA ( 2015 )

4. Hassanzadeh , P.: FindSampo: A Citizen Science Platform for Archaeological Finds on the Semantic Web . Master's thesis , Aalto University, School of Science, Finland (December 2019 ), http://urn. /URN:NBN: :aalto- 201912226669

5. Hyvonen, E., Viljanen , K. , Tuominen , J. , Seppala, K. : Building a national semantic web ontology and ontology service infrastructure|the FinnONTO approach . In: Proceedings of the 5th European Semantic Web Conference (ESWC 2008 ). Springer{Verlag ( 2008 )

6. Hyvonen, E.: Linked Open Data Infrastructure for Digital Humanities in Finland . In: Digital Humanities in Nordic Countries, 5th Conference (DHN 2020 ), Proceedings, accepted. CEUR WS Proceedings ( 2020 ), https://seco.cs.aalto. /publications/2020/hyvonen-lodi4dh-dhn- 2020 .pdf

7. Hyvonen, E.: \Sampo" model and semantic portals for digital humanities on the Semantic Web . In: Digital Humanities in Nordic Countries, 5th Conference (DHN 2020 ), Proceedings, accepted. CEUR WS Proceedings ( 2020 ), https://seco.cs.aalto. /publications/2020/hyvonen-sampos-dhn- 2020 .pdf

8. Thomas , S. , Wessman , A. , Tuominen , J. , Koho , M. , Ikkala , E. , Hyvonen, E., Rohiola , V. , Salmela , U.: SuALT: Collaborative Research Infrastructure for Archaeological Finds and Public Engagement through Linked Open Data . In: Digital Humanities in the Nordic Countries, 3rd Conference (DHN 2018 ) ( 2018 )

9. Veen , V. : The Netherlands during the Napoleonic Era ( 1794 { 1815 ). Using detector nds to shed light on an under-researched period . In: Methods in Con ict Archaeology, 10th Fields of Con ict Conference . vol. 1 , pp. 19 { 30 . Mashantucket Pequot Museum & Research Center ( September 2018 )

10. Wessman , A. , Thomas , S. , Rohiola , V. , Koho , M. , Ikkala , E. , Tuominen , J. , Hyvonen, E., Kuitunen , J. , Parviainen , H. , Niukkanen , M. : Citizen Science in Archaeology: Developing a Collaborative Web Service for Archaeological Finds in Finland , pp. 337 { 352 . Springer ( June 2019 ). https://doi.org/10.1007/978-3- 030 - 14327-5 23

11. Wessman , A. , Thomas , S. , Rohiola , V. , Kuitunen , J. , Ikkala , E. , Tuominen , J. , Koho , M. , Hyvonen, E.: A Citizen Science Approach to Archaeology: Finnish Archaeological Finds Recording Linked Open Database (SuALT) . In: Digital Humanities in the Nordic Countries 2019 . pp. 469 { 478 . CEUR Workshop Proceedings (May 2019 )