aum 2012 - 2nd Workshop on Augmented User Modeling

The digital world and the physical world become more and more connected: our interaction with computer systems has impact on our activities and experiences and vice versa. This changes the way we use technologies and opens up new opportunities for personalization and adaptation. People blog, post, chat, comment, tweet about things that matter to them: what they had for dinner, what their job activities were, what they thought about a particular television broadcast, et cetera. People also share content about their activities, such as pictures taken at a concert, videos of business meetings, reports on business trips and personal stories. This abundant digital information stream has become an important back channel in our daily lives. We constantly create digital traces about our experiences, which can be valuable sources for personalization.
The time is ripe for developing new adaptation paradigms that exploit sensor data and other digital traces of our activities. Hence, traditional adaptation mechanisms (such as feedback, help, guidance) can be extended to become more effective by taking into account not only the user's experience in the digital world (the conventional user modeling paradigm), but also relevant experience of this user or of similar users in the physical world. The latter approach, which is the focus of this workshop, represents an emerging strand of research in which user models are augmented with real world knowledge to enhance adaptation and personalization.
Digital traces can be attributed to more than one individual: circles of friends, scientific communities or even whole populations can be characterized by the topics they tweet about, or things they comment on. Furthermore, events - for example conferences, local or global disasters, political debates - can be modeled by the streams of digital traces generated around these events, including pictures, comments, discussions and reactions. Technological advancements in various areas - such as location tracking, RFID, opinion mining, social signal processing, media aggregation and retrieval - make it now possible to automate the processing of digital traces to enrich the system's understanding about users' experiences in the physical world. These technological developments bring new opportunities to the user modeling community, and at the same time, open up new technological, social, and ethical challenges.
top
Topics
The AUM workshop aims to create a forum for academic and industrial researchers and practitioners to discuss augmented user modeling from three angles:
User Modeling: methods and techniques for analyzing digital traces to capture, represent and connect user experiences:
What sources of digital traces can be used and are there any social and ethical constraints?
What aspects of user experience are captured in different digital traces and what techniques can be used to analyze the streams of digital traces?
How can digital traces be processed, connected, and aggregated to provide useful information for modeling users and their activities/events?
To what extent do these models represent what people, groups and events 'really' are in the physical world, e.g. do they conform to models and theories from social science?
Alignment: methods and techniques for augmenting user models by aligning heterogenous user data traces:
How can digital traces be connected and represented to augment existing user models and to create holistic models of users, events, objects and groups?
How can parts of these holistic models be identified that are relevant to a user context?
How can different perspectives on activities and events be catered for and should they be aggregated in augmented user models?
Application: personalization and adaptation approaches and application areas which can benefit from augmented user models:
How can adaptation and personalization approaches benefit from augmented user models?
What are the potential application domains (e.g. adaptive simulators, personalized virtual museums, personalized media retrieval, personalized information portals, personal assistants) and how can augmented user modeling improve the user experience in these domains?
Which types of personalization, recommendation and information filtering are possible and desirable for different applications or different real-world events (e.g. entertainment activities, job tasks, breaking news, e-learning)?