=Paper=
{{Paper
|id=None
|storemode=property
|title=Towards Community Browsing for Shared Experiences: The WeBrowse System
|pdfUrl=https://ceur-ws.org/Vol-918/111110201.pdf
|volume=Vol-918
|dblpUrl=https://dblp.org/rec/conf/at/Yee-KingCJOHAPSd12
}}
==Towards Community Browsing for Shared Experiences: The WeBrowse System==
https://ceur-ws.org/Vol-918/111110201.pdf
Towards Community Browsing for Shared Experiences:
The WeBrowse System?
Matthew Yee-King3 , Roberto Confalonieri1 , Dave de Jonge2 , Nardine Osman2 , Katina
Hazelden3 , Leila Amgoud1 , Henri Prade1 , Carles Sierra2 , and Mark d’Inverno3
1
Institut de Recherche en Informatique Toulouse (IRIT)
Universitè Paul Sabatier
118 Route de Narbonne
31062 Toulouse Cedex 9, France
{roberto.confalonieri,amgoud,prade}@irit.fr
2
Articial Intelligence Research Institute (IIIA-CSIC)
Campus de la Universitat Autònoma de Barcelona
08193 Bellaterra, Catalonia, Spain
{davedejonge,sierra,nardine}@iiia.csic.es
3
Dept. of Computing, Goldsmiths, University of London,
London SE14 6NW, United Kingdom
{dinverno,exs01kh2,mas01mjy}@gold.ac.uk
Abstract. We introduce the new concept of community browsing: a group of
people browsing the web together and simultaneously. Community browsing is
part of the broader notion of shared experience, where individuals share the ex-
perience of an event. We have developed a prototype of a mobile application that
enables community browsing, and involves new technologies such as a peer-to-
peer Electronic Institution and bipolar preference aggregation.
1 Introduction
Cultural institutions such as museums have been placed under financial pressure by
the current European economic crisis [6], making it more difficult to physically visit
them. It is therefore interesting to see whether we can exploit mobile devices and social
networks to enrich and encourage the experience of cultural artifacts online.
In this paper we report on ongoing work on enabling community browsing. We
describe an application for community browsing that uses intelligent agents to aggregate
the preferences of the individual users into community decisions, and an electronic
institution to enforce the norms of the community onto the agents.
2 The WeBrowse Application
Community browsing means that members of a community browse the web simulta-
neously in a synchronised way. All users see the same things at the same time and
know they are doing so. Actions impacting the community need to be agreed upon
?
AT2012, 15-16 October 2012, Dubrovnik, Croatia. Copyright held by the author(s).
by the community members and individual decisions are aggregated into community
decisions.
We have developed an iOS application called WeBrowse that allows community
browsing of an image database. Each member of the community starts an instance of
WeBrowse on its own mobile device and joins the community session. During the ses-
sion images from a database are displayed to the users, one by one. All users always see
the same image at the same time. Each image is accompanied by a set of tags describing
the image. Every user can express his opinion about the displayed image, by indicating
for each tag whether he likes it or not. An intelligent software agent then collects the
opinions of every user for every tag of the image, and aggregates them into a ‘collective
opinion’ about the image. This information is then used by the database to select the
next image to display.
We have implemented WeBrowse around an Electronic Institution (EI) [1, 5]. Elec-
tronic Institutions make it possible to develop programs according to a new paradigm,
in which tasks are executed by independent agents. The EI ensures that the agents be-
have according to the norms that are defined for the application. A difference with the
existing work on EI however, is that our EI is implemented as a peer-to-peer system
rather than a centralized system.
In order to aggregate the opinions of the users and determine which image to display
next, we apply a possibilistic bipolar representation model as described in [2, 3]. This
aggregation function assigns a weight to each tag associated with the current image.
These weights are then used to rank the images in the database, according to their tags.
The image with highest rank is returned as the next image to the community [4].
We have performed some initial user trials with the application, which indicated that
the system indeed enhances the social experience of its users.
3 Acknowledgments
This work is supported by CHIST-ERA project ACE.
References
1. Arcos, J.L., Esteva, M., Noriega, P., Rodrı́guez-Aguilar, J.A., Sierra, C.: Engineering open
environments with electronic institutions. Engineering Applications of Artificial Intelligence
18(2), 191–204 (2005)
2. Benferhat, S., Dubois, D., Kaci, S., Prade, H.: Bipolar possibility theory in preference model-
ing: Representation, fusion and optimal solutions. Information Fusion 7(1), 135–150 (2006)
3. Casali, A., Godo, L., Sierra, C.: A graded BDI agent model to represent and reason about
preferences. Artificial Intelligence 175(7-8), 1468–1478 (2011)
4. De Calmès, M., Dubois, D., Hullermeier, E., Prade, H., Sedes, F.: Flexibility and fuzzy case-
based evaluation in querying: an illustration in an experimental setting. Int. Journal Uncer-
tainty, Fuzziness and Knowledge-Based Systems 11(1), 43–66 (2003)
5. d’Inverno, M., Luck, M., Noriega, P., Rodrı́guez-Aguilar, J.A., Sierra, C.: Communicating
open systems. Artificial Intelligence (In Print, 2012)
6. Newman, K., Tourle, P.: The Impact of cuts on UK museums A report for the Museums As-
sociation (July) (2011)