Sungrim Kim

Multicontext-Aware recommendation for ubiquitous commerce

Recommender systems address a variety of ubiquitous commerce needs. In ubiquitous commerce, contextual information must be incorporated into the recommendation process. The total amount of information is larger due to the greater number of contexts in multicontext environments. Multicontext therefore requires a more accurate and rapid recommendation method. This paper proposes a multicontext-aware recommendation for ubiquitous commerce using consumers preferences and behavior as a weighting factor. The recommendation method is described and an application prototype is presented. Several experiments are performed and the results verify that the proposed methods recommendation performance is better than other existing method.

Just-in-Time Recommendation Using Multi-agents for Context-Awareness in Ubiquitous Computing Environment

Ubiquitous computing is a vision of future in which a computing technology becomes pervasive. One of the most important attributes in ubiquitous computing is context- awareness. The application that the user’s contexts are changed continuously over time normally requires a just-in-time recommendation. This paper describes a new just-in-time recommendation method using multi-agents for context-awareness in ubiquitous computing environment. For the just-in-time recommendation, we locally store the recommendation information that the user is likely to need in the near future based on predictions. To predict the information that be expected to be needed by the user, we use data mining methods. We propose a JITR Server Context Agent and a JITR Client Context Agent. They make the just-in-time recommendation continuously with the change of user’s contexts and solve the limitation of storage. We explain our method with an example. Several experiments are performed and the experimental results show that our method has a good recommendation performance.

Proactive Information Service through User Participation in Context-aware Application

User participation is an essential aspect of Web 2.0. Web 2.0 creates a context-aware interactive environment that supports the participation of users to enhance information service. Tagging is a key participation method by which users can enrich information in Web 2.0. The authors present a proactive information service through user participation in context-aware application. The proposed method uses participation information acquired in users contexts. Using user participation from contexts, the method provides useful information in a timely manner. The information service considers contexts as the tag and provides information that may be useful without requiring any action on the part of the user.

Attention-Based information composition for multicontext-aware recommendation in ubiquitous computing

The recommender system that offers useful information to the users becomes more important in ubiquitous computing. Context-aware recommender systems are a core component in ubiquitous application. Because of the greater number of contexts, the total amount of information is larger in multicontext-aware environments. Multicontext therefore requires a more efficient recommendation method. We propose an attention-based information composition for multicontext-aware recommendation. We explain that the recommendation process would be comprised of three steps. Our proposed method is capable of recommending accurately and rapidly.

Context-Aware recommendation service using multi-leveled information in mobile commerce

Recommender systems are being used by an ever-increasing number of electronic commerce applications to help consumers find products to purchase. Recommender systems in mobile commerce should be context-aware to understand each consumers contexts anywhere at anytime. This paper proposes a new context-aware recommendation service that enables consumers to obtain relevant information efficiently by using multi-leveled information. This paper describes the recommendation method and presents application scenario that utilizes the method. Several experiments are performed and the results verify that the proposed methods recommendation performance is better than other existing methods.

Ubiquitous information retrieval using multi-level characteristics

Applications of ubiquitous computing are increasingly leveraging contextual information from several sources to provide users with behavior appropriate to their environment. The method of information retrieval is one of the most fundamental research issues in ubiquitous computing. Applications where users contexts change continuously over time require prompt retrieval of relevant information. This paper proposes a new ubiquitous retrieval method that enables users to obtain relevant information efficiently using the multi level characteristics of the contexts. This paper describes the ubiquitous information retrieval process. Several experiments are performed and the results verify that proposed method has better retrieval performance.