Espen Uleberg, Museum of Cultural History Oslo

Keywords: Exploratory Data Analyses; Museum Database; Norway

Introduction
It has for a long time been acknowledged that digital representation of archaeological objects will open up new venues for research. Llobera (2010) referred to how archaeological information can be represented, captured, and analysed through the use of information systems. Although many share this opinion, the realization of this potential is slowly being realised. One reason for this is that large datasets have relatively recently been made accessible for wider audiences. A long-lasting cooperation among the university museums in Norway has provided extensive datasets that are freely available. This paper will present the Norwegian MUSIT infrastructure and discuss some aspects and possibilities that this infrastructure affords.
MUSIT – a national infrastructure
The Norwegian university museums cooperate in the organization MUSIT (Museum IT). MUSIT creates a national infrastructure for collections for natural history and cultural history. This cooperation is a continuation of extensive projects that started in the 1990s. The main parts of the present system was developed in the early 2000s, and MUSIT is now in the process of moving to a new IT-architecture that will have a web-based user interface.
The information that is available today is the result of the extensive digitalization work that was carried out in the 1990s and the fact that all cataloguing has been done in this system since 2006. The media database in MUSIT is in a certain extent integrated with archaeological, numismatic and ethnographic artefacts, while the different artefact collections are not interlinked. The new IT-architecture will improve the integration of images and artefacts, and especially the coin collection will be well integrated with the archaeological material.
The metadata from MUSIT is published at www.unimus.no, and as of today well over 1.4 million database entries have been published. Several thousand objects are published with an image, most of them with a CC BY-SA license. In addition, almost all entries are georeferenced. This means that anyone can use this material on a webpage or download a dataset and create maps.
The work in MUSIT will follow the FAIR-principles for data publication; the data should be Findable, Accessible, Inter-operable, and Retrievable (https://www.go-fair.org/fair-principles/). Several of the images and archaeological objects are linked to the Norwegian Sites and Monuments register, Askeladden, and Askeladden (www.kulturminnesok.no)and MUSIT link to each other on their web pages.

ADED (Archaeological Digital Excavation Documentation)
The Museum of Cultural History at the University of Oslo is leading the ADED project (Archaeological Digital Excavation Documentation) which will provide a further basis for integrating archaeological information. The ADED project will merge digital excavation documentation from single projects and provide a common interface where metadata can be queried across single projects.
All of the Norwegian university museums have since 2011 used the same system, Intrasis (https://www.intrasis.com/), for excavation documentation. Intrasis can be used as a very flexible system, but the museums in Norway, thinking of a closer integration in the MUSIT system, decided to use common templates for their excavations. This has provided a solid ground for the work in ADED. ADED will become an integrated module in the MUSIT system, and in this way artefacts will be linked to their detailed context information. MUSIT also provides link to images, and the system will also include links to excavation reports.

Exploratory Data Analyses
MUSITs datasets that can be downloaded from unimus.no can be a good basis for Map based Exploratory Data Analysis (EDA) (Andrienko and Andrienko 2006, Uleberg and Matsumoto 2016). A main idea with EDA is to work iteratively. The initial observations are used in analyses, and each iteration will use the results as observations that give ideas to constellations that will used in new analyses. The data can be grouped, divided and viewed in a range of different scales that can lead to new understanding and new sets of questions.
The MUSIT data is a quite large georeferenced dataset that can form the basis for such analytic work. As with all large datasets, it is necessary to complement the initial query results with extensive research to validate the results. In a museum collection, the precision of the context information is very variable, and the precision level of the coordinates is an important metadata element. It is possible to argue that minor errors can be ignored in a large dataset, but anyhow one should always evaluate the results in a proper manner.
One example of working with data sets from MUSIT in an EDA context can be found in the project Dynamic Distributions (Matsumoto and Uleberg, Uleberg and Matsumoto 2016). In this project, Stone Age artefacts are combined with a landscape reference system (Puschmann 1998) for Norway. The artefacts are grouped at different temporal scales, and related to different landscape types. The Puschmanns landscape types are used as defined or grouped into larger units. The coordinate precision is one of the criteria for including artefacts in the analyses. More artefacts can be included at courser levels while only those with a high coordinate precision can be included at the more detailed levels.
Concluding remarks
The MUSIT infrastructure creates a basis for new venues of archaeological research. The coming results from the ADED project will to an even greater extent give opportunities to look for connections and patterns that can give new insight. It will be possible to send less time on data collection, and more time on analysing data at different scale levels and in combinations with different types of datasets. The most challenging can be to find the right questions, and the mapbased Exploratory Data Analyses gives many opportunities to find them.

Litterature:
ANDRIENKO, N. and ANDRIENKO, G. (2006) Exploratory Analysis of Spatial and Temporal Data. Berlin, Heidelberg: Springer Verlag.
Llobera, Marcos 2011, Archaeological Visualization: Towards an Archaeological Information Science (AISc). Journal of Archaeological Method and Theory 18(3):193-223
PUSCHMANN, O. (1998) The Norwegian landscape reference system – use of different sources as a base to describe landscape regions. NIJOS report 12/98. Ås: NIJOS.
Matsumoto, M. and Uleberg, E. (2015a) ‘Scale dependentatterns in South–Eastern Norway’, in Traviglia, A. (ed.) Across Space and Time. Papers from the 41st Conference on Computer Applications and Quantitative Methods in Archaeology. Perth, 25–28 March 2013. Amsterdam: Amsterdam University Press, pp. 283–288.

Uleberg, E. and Matsumoto, M. (2015) ‘Visualization based on the Norwegian University Museum Database’, in Gilligny, F., Djindjian, F., Costa, L., Monscati, P., and Robert, S. (eds) CAA2014. 21st Century Archaeology. Concepts, Methods and Tools. Proceedings of the 42nd Annual Conference on Computer Applications and Quantitative Methods in Archaeology. Oxford: Archaeopress, pp. 323–331.

Uleberg, E. and Matsumoto, M. (2016) ‘Dynamic distributions in macro and micro perspective’,
in Stefano, C., Scopigno, R., Carpentiero, G., and Cirillo, M. (eds) CAA2015. Keep the Revolution Going >>> Proceedings of the 43rd Annual Conference on Computer Applications and Quantitative Methods in Archaeology (Vol. 1). Oxford: Archaeopress, pp. 309–317.