A Complex Network Approach for Museum Services

A Model for Digital Content Management

Filippo Eros Pani, Simone Porru, Matteo Orrù and Simona Ibba

Department of Electrical and Electronic Engineering, University of Cagliari, Piazza d'Armi, Cagliari, Italy

Keywords: Knowledge Management, Complex Network, Knowledge Base, Museum Services.

Abstract: In a globalized economy, cultural heritage is a strong attractor. Thanks to ICT, it is possible to trigger new

development dynamics. For cultural heritage, the contribution of new technologies can offer the highest

degree of distribution and access opportunities. A modern museum can actually be seen as a complex ICT

system, deeply interconnected, with typically a large quantity of data to manage, extremely dynamic due to

ever-changing temporary exhibitions, and with applications that feature a high level of usability for a higher

visitor involvement. The aim of this position paper is to create an approach geared to provide aggregated

information on the nature, range and articulation of the belongings of the museum, through a paradigm based

on the concept of complex network. Applying the complex network model, it will be possible to map a corpora

of items made of works of art, artifacts and any object of interest for a museum. The implications derived

from the adoption of this approach are multifarious: for example, a curator could evaluate partnership

opportunities in the organization of temporary exhibitions, guided paths or catalog editing through an analysis

of the relations between the items in their museum and the ones in other museums.

1 INTRODUCTION

The cultural heritage sector has an enormous potential

for innovation, with astounding development

prospects. This evolution scenario is of special

interest to countries owning a rich cultural heritage.

Such a wealth can also be relevant as an economic

asset, with very interesting and varied consequences

and characteristics.

Museums are known for the large quantity of data

they need to manage, and show an extreme

dynamism, for example due to temporary exhibitions.

Its applications must possess a high level of

“usability” so that a higher number of visitors can be

involved in the most “intense” way possible. It is

implied that the use of digital technologies in the

museum sector offers many opportunities for

innovation (Pani et al., 2014). Thanks to the use of the

Internet and to the new forms of interaction with

information systems, the ways in which museums can

be enjoyed are expanding, for information and

communication management systems as well as for

increasingly “bold” opportunities leading to a new

type of museum, namely an interactive one (Kelly,

2014) (Sahba, 2014) (Chianese and Piccialli, 2014).

The aim of our proposal is to provide aggregated

information on the nature, the extension and the

articulation of the belongings of museums. The

information related to each artifact needs to be

managed through a paradigm based on the concept of

complex network, integrating “punctual” information

(as related to a single item, like in a traditional

catalog) with another type of information related to

the connections among the different items that are

hosted by the museum (gallery, exhibition, etc.). This

second type of information is of a “topological”

nature.

The paper is structured as follows: Section Two

shows an overview of the complex networks and in

Section Three we propose our approach In Section

Four some possible applications to museum services

in real world scenarios are presented. Lastly, Section

Five includes the conclusions and some reasonings

about our work.

2 RELATED WORK

In the recent years there has been a surge of interest

and an increasing number of studies on complex

networks (Newman, 2003).These systems are found

in natural sciences, for example in proteomics and in

neurophysiology, in social sciences (i. e., social

216

Pani, F., Porru, S., Orrù, M. and Ibba, S..

A Complex Network Approach for Museum Services - A Model for Digital Content Management.

In Proceedings of the 7th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2015) - Volume 3: KMIS, pages 216-221

ISBN: 978-989-758-158-8

networks like Facebook), in technology and computer

science (i.e., Internet and the Web). The increasing

number of scientific works published on prestigious

journals such as Nature, Science and PNAS shows

that the subject has become one of the most important

in statistical physics and science in general. We live

in an age where information is the ultimate source of

meaning, as well as economic value, and the

information is propagated and is processed through

their complex networks.

It has been found that networks of different nature

share some properties (Valverde et al., 2002)

(Valverde and Sole, 2003) (Concas et al., 2013).

Many different kind of complex networks have been

proved to present the small-world property. Reported

for the first time by Milgram (Milgram, 1967) it has

been then popularized as “six degree of separation”.

This networks present small values for average

shortest paths and high value of the clustering

coefficient. This implies that network nodes are

highly connected and that the number of “hops” that

separates two pairs of nodes are on average small if

compared to a random network of the same size. As a

result it is possible to find a path that connects two

nodes with a relatively small number of edges, but

also to change a path in case of edges being removed.

In general, in complex networks, the distribution

of the nodes' in (out) degree is represented by an

exponential function. This kind of network usually

presents few nodes that are highly connected, with a

far greater number of connections, whereas, on the

contrary, the vast majority of nodes present just few

connections with the rest of the network. This

property has practical implications: for example if

one considers the reliability of a network, that results

robust to the random failure but vulnerable to failure

involving the highly connected nodes or hubs.

One of the most interesting property of complex

networks is that they present a community structure

(Girvan and Newman, 2001), (Fortunato, 2010).

Inside a network, a community is a subnetwork whose

nodes present denser connections to each other, if

compared to the connection with nodes outside the

community. Researchers of different background

tackled the problem of determining the best division

in communities for a network, using different

approaches (statistical, algorithmical, etc). Finding a

community structure implies finding the elements

that present the highest interaction, meaning that they

likely belong to the same social group or share some

meaningful features.

Recently some scholars began to study the

hierarchical aspects of complex network. The

Louvain algorithm, as a matter of fact, is already a

multilayer algorithm. Clauset et al. (2007, 2008)

proposed a method to find the hierarchical structure

of a network using a Monte-Carlo Markov Chain

approach. The hierarchical structure retrieved has

been shown to be useful to detect missing links

between nodes.

Future applications of the use of complex

networks in museums include producing and

providing visitors with customized navigation

environments, capable of leveraging the nowadays

ubiquitous handheld devices. The interest on the use

of mobile devices as a means for enhancing users'

experience in museums has been increasing lately

(Rubino et al., 2013). In the work of Rubino, mobile

interactive guides, tailored for children and adults, are

the final product of the proposed MusA framework.

This framework also leverages a vision-based indoor

positioning system, capable of providing location

based services such as customized visit paths.

Several articles has also been produced on visitor

circulation patterns (Bitgood, 2006), often focusing

on the benefit/cost ratio as the determing factor in

visitor's path choices. While some general visitors'

preferences are not linked to the exhibit's quality

and/or typology (e.g., right-hand side circulation,

preference for limited choice points instead of

multiple choice, one-sided visiting pattern in two-

sided exhibitions), others can be heavily dependent

on the differences between perceived exhibits' values.

Thus, understanding how movement patterns change

according to different exhibits positioning can be

regarded as an effective way to reveal the items'

perceived value.

Providing visitors with a customized navigation

environment is also a primary goal for recommender

systems. In the case of cultural sites, especially when

they contatin Points Of Interests (POI), classical

recommendation techniques have been used in

conjunction with context-aware services to provide

users with a dynamic visiting path, which therefore

changes according to the mutable needs of the users

(Bartolini et al., 2014).

The exploitation of collaboration between

museums, and thus metadata standardization, is also

another topic of interest that is regarded as a major

factor in producing disruptive applications in the

cultural heritage scenario (Skinner, 2014). To this

respect, leveraging smartphones and Web 2.0

technologies can represent a means to detect whicth

context-aware metadata can be used for standardizing

user-generated knowledge (Flanagan and Carini,

2012).

A Complex Network Approach for Museum Services - A Model for Digital Content Management

217

3 THE PROPOSED APPROACH

The proposed approach is based on the concept of

complex network (or graph), that is a structure with

two main separate elements: nodes, which represent

the basic elements of the graph, and node

connections, called branches or arches. An example

of a network we face every day is the road network.

We could associate cities to nodes, and roads

connecting cities to branches. The complexity is

intrinsic to the networks as their size increases, that is

to the number of involved elements (nodes and

branches), to the point that it becomes extremely

complex to understand its structure, its behavior, and

its evolution. For this reason, in order to study

complex networks, tools, methods, and algorithms

coming from a multidisciplinary knowledge corpus

(statistical physics, sociology, etc.) are used, a corpus

that has been gradually built over the years.What

makes complex networks interesting is the fact that

they constitute a mathematical model that can

represent facets and artifacts of human life, and

several natural phenomena. Networks of different

origin and nature (from electric networks to metabolic

ones) possess the same properties. It is possible to

apply the same concept of network to human

knowledge. In this case, the term knowledge network

is used. An example of knowledge network is

represented by bibliographic networks created from

author collaboration (co-authorship) information or

quotations appeared in reference sections.

Figure 1: Collaboration network of jazz musicians. Each

node is a Jazz musician and an edge denotes that two

musicians have played together in a band (Gleiser and

Danon, 2003). Different colors represent the community

retrieved with the FastGreedy Algorithm (Clauset et al.

2004).

4 MUSEUM SERVICES

It is possible to apply the complex network model to

map a corpus of items, along the same lines as

bibliometric studies, but which is larger and more

varied than bibliographic heritage, including works of

art, artifacts, and any items of interest to a museum.

The way howartifacts can be associated to each other

on the basis of some of their features is actually

intuitive. For example, sets of artifactscan be

considered connected if they belong to the same

artistic movement, or to the same historical period, or

school, or location, author, type, subject, etc.

The implications derived from such an approach

are multifarious for curators. Curators could, for

example, analyze the nature of the relations between

the items in their museums and items belonging to

other museums, and evaluate potential partnerships

when organizing temporary events, guided paths,

creating catalogs.

Starting from information included in catalogs, a

complex network with items and their relations

(aggregated and at different granularity levels) will be

created. It will include a visual representation, in a

synthetic form, explorable at a glance, and it will

allow to investigate the complexity of the relations

between items.

That network is, in fact, a mathematical model on

which it will be possible to apply different algorithms

to the collection of network metrics and perform

statistical analysis. It will be possible to, for example,

obtain information on the nature and strength of the

connections between items through the use of

clustering or community-detection algorithms. It will

be possible to follow the evolution of the items over

time, or associate information on the items to

information of another nature (for example,

pertaining to location of origin). It will be also

possible to filter the elements of the network

according to several parameters (period, artistic

school, author, etc.), to analyze only the subnetworks

of relevant elements.

The enhanced capabilities of museum resources as

parts of a complex network could also be leveraged

for enhancing on-site visitors' experience, providing

them with different alternative paths to fully enjoy

their visit. For instance, a visitor could choose among

a “chronological”, “paintings by artist”, or a

“stylistic” path, by simply selecting their choice

through a dedicated mobile application, specifically

designed to work with the complex network

associated to the resources in the exhibition. This

could also mean that, by studying the visitors'

preferences with regard to the different visiting paths

(performed through the analysis of the data collected

via the mobile app), it will also be possible to place

the resources according to the most popular path (e.

KMIS 2015 - 7th International Conference on Knowledge Management and Information Sharing

218

g., visitors might be more interested in seeing a

certain artist's paintings first, then ones from another,

and so on, in a “paintings by artist” fashion).

In Fig. 2 an explicative diagram illustrating the

interaction between a visitor and a dedicated mobile

app is reported, so as to provide an example of how

new information could be added in a complex

network for resources in a museum. According to the

proposed diagram, a visitor might use the museum

application to explore the suggested predefined

visiting paths, that is, paths where museum resources

(e. g., paintings, sculptures) are shown to the visitor

in a predefined order, such as a chronological order,

or a “by artist” order.

Since the museum app communicates with a server,

the visitor's choice will be recorded on the system.

Thanks to the recording of such information, it will

be possible to see which kind of suggested path is the

most popular among visitors and, observing their

behavior after that choice (i. e., if they followed that

path or decided to leave it very soon during their

visit), to check if it seems to satisfy the user's needs

and meet their expectations. Each deviation from the

chosen suggested path could also show the

preferences of common visitors more clearly. For

example, if many visitors show interest toward a

particular piece of art, and read its description (e. g.,

through a QR-code placed next to the item, which

upon acquiring it will send a request to the server for

obtaining the object's information) instead of

following the predefined path, it could be useful to

insert the said piece of art in a list of “suggested

items” among those next to the visitor.

Figure 2: A possible interaction between visitors and a dedicated mobile app.

A Complex Network Approach for Museum Services - A Model for Digital Content Management

219

following the predefined path, it could be useful to

insert the said piece of art in a list of “suggested

items” among those next to the visitor.

After collecting enough data concerning the order

in which items were actually seen by the visitors, and

which suggestions they were prone to accept the

most, it will be possible to enrich the information

contained in the resources' complex network, and

then design a so-called “optimal path” for visiting the

museum. The museum could then easily adapt to its

visitors' preferences, that is, learn from previous

experiences.

5 CONCLUSIONS

In this paper we proposed an innovative approach to

exploit the resources hosted on a museums and the

like, by leveraging ITC technologies on one side, and

a powerful statistical model called complex network

on the other. We described several ways in which this

integration might take place.

The enhanced capabilities of museum resources as

parts of a complex network could be leveraged for

enhancing on-site visitors' experience, providing

them with different alternative paths to fully enjoy

their visit. For instance, a visitor could choose among

a “chronological”, “paintings by artist”, or a

“stylistic” path, by simply selecting his/her choice

through a dedicated mobile application, specifically

designed to work with the complex network

associated to the resources in the exhibition. This

could also mean that, by studying the visitors'

preferences with regard to the different visiting paths

(performed through the analysis of the data collected

via the mobile apps), it will also be possible to place

the resources accordingly to the most popular paths

(e. g., visitors might be more interested in seeing

paintings first from an artist, than from another, and

so on, in a “paintings by artist” fashion). As indoor

localization services are currently a trending topic

(Mighali et al., 2015) in the tourism industry,

succeeding in leveraging the resources which a

museum holds for improving cultural experience and

indoor localization technologies represents a fairly

attractive opportunity for both research and industry.

Museum resources could also be described

through metadata associated to visitors' paths

preferences, which will be used for characterizing a

binary relationship between resources (e. g., two

paintings could be linked together when most of the

visitors prefer to see them one after the other), thus

revealing connections that were previously hidden or

not apparent.

Studying the relationship between path

preferences and the so-called 'hypercongestion' (i. e.,

the number of visitors exceeds musem's physical

space capacity) will also serve as an indicator of how

visitors react to a more crowded environment, as a

similar study on this subject (Yoshimura et al., 2014),

conducted on the Louvre museum, has tried to

understand.

REFERENCES

Bartolini, I., Moscato, V., Pensa, R. G., Penta, A.,

Picariello, A., Sansone, C. and Sapino, M. L. (2014).

Recommending multimedia visiting paths in cultural

heritage applications. In Multimedia Tools and

Applications, pp. 1-30, DOI: 10.1007/s11042-014-

2062-7.

Bitgood, S. (2006). An analysis of visitor circulation:

Movement patterns and the general value principle. In

The Museum Journal, Vol. 49, Issue 4, pp. 463-475.

Chianese, A., and Piccialli, F. (2014). Designing a smart

museum: when Cultural Heritage joins IoT. In Eighth

International Conference on Next Generation Mobile

Apps, Services and Technologies (NGMAST), IEEE, pp.

300-306.

Clauset, A., Moore, C. and Newman, M. E. J. (2007).

Structural Inference of Hierarchies in Networks. In

Airoldi, E. M. et al. (Eds.): ICML 2006 Ws, Lecture

Notes in Computer Science 4503, pp. 1-13. Springer-

Verlag, Berlin Heidelberg.

Clauset, A., Moore, C. and Newman, M.E.J. (2008).

Hierarchical structure and the prediction of missing

links in networks. In Nature 453, pp. 98-101.

Concas, G., Monni, C., Orrù, M. and Tonelli, R. (2013). A

study of the community structure of a complex software

network. In Proceedings of the 2013 ICSE Workshop

on Emerging Trends in Software Metrics, WETSoM

’13, San Francisco, CA, USA, pp. 14-20.

Flanagan, M. and Carini, P. (2012). How games can help us

access and understand archival images. American

Archivist, Vol. 75, Issue 2, pp. 514-537.

Fortunato, S. (2010).Community detection in graphs. In

Physics Report, 486:75-174.

Girvan, M. and Newman, M. E. J. (2001). Community

structure in social and biological networks. In Proc.

Natl. Acad. Sci. U. S. A., 99(cond-mat/0112110):8271–

8276, 8 p.

Gleiser, P. M. and Danon, L. (2003). Community structure

in jazz. In Advances in Complex Systems, 6(4):565-573.

Kelly, L. (2014). The Connected Museum in the World of

Social Media. In Museum Communication and Social

Media: The Connected Museum. Ed. Kirsten Drotner

and Kim Christian Schrøder. New York-London:

Routledge, pp. 54-71.

KONECT, May 2015. Jazz musicians network dataset -

KONECT.

Mighali, V., Del Fiore, G., Patrono, L., Mainetti, L., Alletto,

S., Serra, G., andCucchiara, R. (2015). Innovative IoT-

KMIS 2015 - 7th International Conference on Knowledge Management and Information Sharing

220

aware Services for a Smart Museum. In Proceedings of

the 24th International Conference on World Wide Web

Companion, pp. 547-550). International World Wide

Web Conferences Steering Committee.

Milgram, S., The small world problem (1967).

InPsychology Today 2, pp. 60-67.

Newman, M. E. J. (2003). The structure and function of

complex networks. In SIAM REVIEW, 45:167-256.

Pani, F. E., Concas, G., Porru, S., (2014). An Approach to

Multimedia Content Management. In Proceedings of

the 6th International Conference on Knowledge

Engineering and Ontology Development, KEOD 2014,

21- 24 October, Rome, Italy.

Rubino, I., Xhembulla, J., Martina, A., Bottino, A., and

Malnati, G. (2013). MusA: using indoor positioning

and navigation to enhance cultural experiences in a

museum. In SENSORS, Vol. 13, Issue 12, pp. 17445-

17471. ISSN: 1424-8220.

Sahba, F. (2014). Museum automation with RFID. In World

Automation Congress (WAC), IEEE, pp 19-22.

Skinner, J. (2014). Metadata in Archival and Cultural

Heritage Settings: A Review of the Literature. In

Journal of Library Metadata, Vol. 14, Issue 1, pp. 52-

68. DOI: 10.1080/19386389.2014.891892.

Sahba, F. (2014). Museum automation with RFID. In World

Automation Congress (WAC), IEEE, pp 19-22.

Valverde, S., Cancho, R. and Sole, R. V. (2002). Scale free

networks from optimal design. In Europhysics Letters,

60, 2002.

Valverde, S. and Sole, R. V. (2003). Hierarchical small

worlds in software architecture. arXiv:cond-

mat/0307278v2.

Yoshimura, Y., Sobolevsky, S., Ratti, C., Girardin, F.,

Carrascal, J. P., Blat, J., & Sinatra, R..(2014). An

analysis of visitors’ behavior in The Louvre Museum: a

study using Bluetooth data. In Environment and

Planning B: Planning and Design, 41(6), pp.1113-

1131.

A Complex Network Approach for Museum Services - A Model for Digital Content Management

221