PersDB 2010

Abstract. In this paper a new formal approach is proposed to retrieval personalization, based on both a query personalization process at the client’s side and a light document adaptation at the information server’s side. The proposed solution relies on the use of a domain ontology: queries and documents are in fact indexed by sets of concepts. The query personalization process is finalised to clarify what we call the central query concepts based on the importance of linked concepts in the considered ontology. The initial query as well as its clarifications are sent to the server, which revises the document representations based on both the query and the concepts clarifications. The proposed solution does not require that the information server maintains any user profile.

Abstract. Relational database systems are becoming increasingly popular in the scientific community to support the interactive exploration of large volumes of data. In this scenario, users employ a query interface (typically, a web-based client) to issue a series of SQL queries that aim to analyze the data and mine it for interesting information. First-time users, however, may not have the necessary knowledge to know where to start their exploration. Other times, users may simply overlook queries that retrieve important information. In this work we describe a framework to assist non-expert users by providing personalized query recommendations. The querying behavior of the active user is represented by a set of query fragments, which are then used to identify similar query fragments in the recorded sessions of other users. The identified fragments are then transformed to interesting queries that are recommended to the active user. An experimental evaluation using real user traces shows that the generated recommendations can achieve high accuracy.

Abstract. Web service discovery aims at finding available services that match a given service description. This involves mainly the matchmaking of the functional parameters of the services, whereas non-functional attributes can also be considered and aggregated in the matching score of a candidate ser- vice as additional criteria for ranking the results. In this paper, we address the problem of re-ranking discovered ser- vices that include nominal attributes in their descriptions in order to satisfy users with diverse preferences. We present an approach to diversify the search results combining the degree of match on functional parameters with a method to achieve good coverage with respect to the values of nominal attributes. An evaluation on a publicly available dataset of Semantic Web services is also presented.

Abstract. The research described in this paper focuses on adapting generic Hypermedia Environments with the use of Semantic Web technologies. The use of machine understandable semantics in Web applications is increasing significantly, ranging from semantic search queries to personalized product recommendations. Given the significance of human factors in Web structures’ transformations, we propose an Ontological Web Personalization and Adaptation Mechanism based on users’ cognitive parameters. This mechanism consists of a number of interrelated components; modeling users based on their cognitive parameters and providing content adaptation based on semantics. An RDFa schema has been designed that enables standard annotations in any XHTML Web-page, thus making structured data available for the adaptation process, but also for any service or tool that supports the same standard. Our work has been positively evaluated using an existing commercial Web-site that was filtered through the adaptation mechanism based on users’ cognitive parameters.

Abstract. Over the last few years, computational power, storage capacity, and sensing capabilities of mobile devices have significantly improved. As a consequence, they have undergone a rapid development from pure telecommunication devices to small and ubiquitous computing platforms. Most importantly, these devices are able to host context-aware applications, i.e., applications that are automatically adjusted to the current context of their user. This, in turn, requires sensing support on the device, the possibility to store context information, and to efficiently access this context information for the automated adaptation of applications. In this paper, we introduce a benchmark for context management in mobile context-aware applications. We present in detail the design and setup of the benchmark, based on an eHealth use case. The benchmark evaluation considers context queries on Android Nexus One cell phones and compares the performance of different settings including relational and objectoriented databases on the mobile device, and an RDF triple store on a stationary computer. The results show significant differences in the settings that have been evaluated and are thus valuable indicators for database selection and system design for mobile context-aware applications.

Abstract. We introduce two criteria for judging "goodness" of the result when combining preference relations in information systems: completeness and consistency. Completeness requires that the result must be the union of all preference relations, while consistency requires that the result must be an acyclic relation. In other words, completeness requires that the result contain all pairs appearing in the preference relations, and only those pairs; while consistency requires that for every pair (x, y) in the result, it must be able to decide which of x and y is preferred to the other. Obviously, when combining preference relations, there is little hope for the result to satisfy both requirements. In this paper, we classify the various methods for combining preference relations, based on the degree to which the result satisfies completeness and consistency. Our results hold independently of the nature of preference relations (quantitative or qualitative); and also independently of the preference elicitation method (i.e. whether the preference relations are obtained by the system using query-log analysis or whether the user states preferences explicitly). Moreover, we assume no constraints whatsoever on the preference relations themselves (such as transitivity, strict ordering and the like).