You are here

Personalization of content and interaction

Project researchers

Personalization technology can effect various aspects of the interaction with users, e.g.: the information selected for presentation; the organization of the overall presentation; the media used to interact with users; the interaction modalities; the flow of the human-system dialogue; the individual vs. group type of adaptation...  Several strands of investigation have been at the top of the i3 team research agenda for many years to study how personalization can improve the user experience to best fit their needs, preferences and contextual features.

Personalization for the cultural heritage sector

The FBK research staff has contributed significantly to the history of ICT technologies applied to the cultural heritage domain, centred around the needs and preferences of visitors and the communicative and educative mission of cultural institutions. From the first pioneer project ALFRESCO (1990-1994) investigating early topics of multimedia presentations in the cultural domain, research efforts evolved into the seminal work of HyperAudio, one of the first context-sensitive, adaptive and mobile guides to museums developed in the late 90s, whose personalization challenges have been the bases for the subsequent European projects HIPS (Esprit #25574) and M-PIRO (IST-1999-10982) and for PEACH, a project funded by the Autonomous Province of Trento, involving international partners. More recently, the research efforts have continued within ITCH (Intelligent Technologies for Cultural Visits and Mobile Education), a joint project between Italy and Israel, sponsored by the Italian Ministry of University and Research and by the Autonomous Province of Trento.

Within the meSch EU funded project we are now investigating more deeply how the process of defining adaptive structures can be improved, made more flexible and portable, by identifying and separating the strength of automatic mechanisms from that of the human judgment and effectively synchronizing the activities of the two actors to achieve a superior quality of the results, a technique effectively used in other cases of human-intelligent systems collaboration and of end-user development.

Personalization for Tangible Embedded and Embodied Interaction

The meSch project allows us to push the boundaries of investigation even further, by uncovering the potential of combining (i) the personalized presentation of digital content with (ii) tangible interaction with technology augmented exhibits or spaces and (iii) social interaction. Indeed, as meSch will bring personalization into smart objects augmented with interaction abilities, how the objects are going to deliver the adapted content according to the context of use needs consideration. The personalization of the interaction is complementary to the personalization of the content, as the (adaptive) narration can be delivered through alternative objects, by different activating actions and social interactions.

Personalization and recommendation services

By definition, recommendation services are functions that exploit information about (i) users’ personal characteristics, preferences and interaction history, (ii) semantic features of item descriptions or information articles and/or evaluation ratings, and (iii) possibly the behavior of a community of other users, to suggest items, pieces of information or services within a large information space, that best suit the user's needs in a given situation and context. Recommenders have reached good maturity and robustness to be deployed in real setting applications in diverse domains and for diverse (context-dependent) user tasks. For this reason, they represent a viable solution to introduce a first personalization layer over content extracted from knowledge sources to tune search results to the interaction context and task. Within meSch we are studying how recommendation algorithms can be adapted to the cultural heritage domain to support both the authoring task of curators and the online information exploration by visitors. Logs from the onsite and online visit experiences are used to guide the recommendation strategies with a hybrid approach that combines content- and collaborative-based methods. In other projects, we have also investigated how social relationships betweens users can be exploited in order to complement recommendations, as in trust-aware recommender systems.

Friday, 18 October, 2013

O. Stock, M. Zancanaro, P. Busetta, C. Callaway, A. Krueger, M. Kruppa, T. Kuflik, E. Not and C. Rocchi "Adaptive, intelligent presentation of information for the museum visitor in PEACH", In UMUAI - User Modeling and User-Adapted Interaction, Volume 17, Number 3, July 2007, 257-304, Winner of the 2007 James Chen Annual Award for Best UMUAI Paper, (published online, 25 April 2007, SpringerLink )

T. Kuflik, O. Stock, M. Zancanaro, A. Gorfinkel, S. Jbara, S. Kats, J. Sheidin, N. Kashtan, A Visitor’s Guide in an Active Museum: Presentations, Communications, and Reflection, in «ACM JOURNAL ON COMPUTING AND CULTURAL HERITAGE», vol. 3, n. 3, 2011 (ACM Digital Library link)

Petrelli D., Ciolfi L., van Dijk D., Hornecker E., Not E., Schmidt A. (2013) Integrating material and digital: a new way for cultural heritage, in «INTERACTIONS», vol. 20, n. 4, July + August 2013, pp. 58 - 63 (ACM Digital Library link)

P. Massa, P. Avesani (2007) Trust-aware recommender systems, In Proceedings of the 2007 ACM conference on Recommender systems, pages 17-24, ACM (pdf)

Research topics: