2nd Workshop on LifeLong User Modelling
نویسندگان
چکیده
Recent technological developments enabled users to create detailed log files containing their everyday activities. These log files, also referred to as lifelogs, open gates for interesting new research challenges, such as the identification of long-term and short-term user interests and behaviour. In the context of this workshop, we focus on an emerging challenge, namely the creation of lifelong user models from a potentially large and diverse data corpus that has been created over a very long period of time. Such lifelong user models pose interesting questions, e.g. regarding the scalability of existing user modeling techniques with respect to the diverse input sources and the pure size of the lifelogs. The importance of this role for the lifelong user model is reflected in the identification of personalised lifelong learning as a Grand Challenge Problem by the Computing Research Association (CRA), United Kingdom Computing Research Committee (UKCRC) and National Academy of Engineering. Following the successful first Workshop on Lifelong User Modelling which was held in conjunction with UMAP 2009 (http://rp-www.cs.usyd. edu.au/~llum/2009_UMAP09_llum/), this workshop provided a platform for researchers from both user modelling and lifelogging communities to discuss emerging research trends in this field.
منابع مشابه
Proceedings of the Lifelong User Modelling Workshop, at UMAP'09 User Modeling Adaptation, and Personalization
This paper explores an architecture for very long term user modelling, based upon personal user model clouds. These ensure that the individual’s applications can access their model whenever it is needed. At the same time, the user can control the use of their user model. So, they can ensure it is accessed only when and where they wish, by applications that they wish. We consider the challenges ...
متن کاملLifelong User Modelling Goals, Issues and Challenges
The combination of low and falling costs of storage, combined with increasingly pervasive and ubiquitous computing are creating the possibility of very long term user modelling. This paper explores some of the fundamental questions about the nature of a lifelong user model. It begins by exploring the relationship between the new goals that lifelong user models can support and the concomitant is...
متن کامل2 nd International Workshop on News Recommendation and Analytics ( NRA 2014 ) In conjunction with 22 nd Conference on User Modelling , Adaptation and Personalization ( UMAP 2014 ) 11 July 2014 , Aalborg , Denmark
The 2nd International Workshop on News Recommendation and Analytics (NRA) brings together researchers on news analytics and stakeholders from the media industry. A particular focus is on news recommender systems, that tailor content from media houses and social sites to the preferences and context of individual readers. The workshop includes one invited speaker from the media industry and five ...
متن کاملEfficient Representation of the Lifelong Web Browsing User Characteristics
Client-based user modelling has already been studied and clearly has its place among generic approaches to the user modelling. It is especially advantageous for lifelong user modelling as it can support the modelling in any time and any place including consideration of user privacy. Emergence of web browser extensions opens up possibilities of pure browser-based realisation of client-based user...
متن کاملA (multi‘domain’sional) Scrutable User Modelling Infrastructure for Enriching Lifelong User Modelling
A Scrutable User Modelling Infrastructure (SUMI) is proposed in this paper, which focuses on three key Lifelong User Modelling (LUM) requirements: interoperability, scrutability and privacy. SUMI attempts to enrich the LUM vision by examining two unexplored “pieces of the puzzle”: the social e-networking and e-commerce domains. SUMI allows users to gather their various (lifelong) models which t...
متن کامل