Context information for knowledge reshaping

Abstract

Abstract:More and more often, we face the necessity of extracting appropriately reshaped knowledge from an integrated representation of the information space. Be such a global representation a central database, a global view of several ones or an ontological representation of an information domain, we face the need to define personalised views for the knowledge stakeholders: single users, companies or applications. We propose exploiting the information usage context within a methodology for context-aware data design, where the notion of context is formally defined together with its role within the process of view building by information tailoring. This paper presents our context model, called the context dimension tree, which plays a fundamental role in tailoring the information space according to user information needs.

CARVE: Context-aware automatic view definition over relational databases

“…The CDT has been formally presented in [10]; it can be specified in several ways, for instance we have implemented both an ontological [20] representation in OWL [37] and an XML representation based on a DTD [21]. In the following we illustrate how, given an application scenario, the possible contexts are built starting from the Context Dimension Tree and its nodes.…”

“…We build on some research already published in other works by the same authors [8][9][10]; the original contributions we present in this paper are: the overall organization of the CARVE methodology, presented step by step; an improved version of the context model we introduced in [10]. In this new definition, we add new elements useful to model and express the aspects necessary to specify the characterization of final users; refinement and correction of the view composition operators [8] and of the strategies to identify the database portion interesting for a given context; their previous versions were tested against the case studies presented in Section 7, and they revealed themselves inadequate to capture all the possible relationships that a designer might be wanting to establish between a context and the associated data.…”

Optimizing Service Selection and Allocation in Situational Computing Applications

“…QoS parameters, however, are subject to high variability. Indeed, the performance of services available in the Internet can vary up to an order of magnitude within the same business day [15], while the user behavior and the characteristics of the pervasive environment (i.e., the application's context [16]) also change at run-time. For example, if a user connected to an ad-hoc WIFI network changes his physical position the network bandwidth might be reduced and one or more physical devices could even become unreachable.…”

Emergent Semantics and Cooperation in Multi-knowledge Communities: the ESTEEM Approach

“…In Esteem, context is described through the Context Dimension Tree (CDT) [9], a context model that has been conceived to support the tailoring of the peer data and services according to the current context. For example, the doctor considered in our application scenario is interested in acquiring information on the diseases and symptoms common in Central Africa and on the available care facilities; in another scenario, a laboratory technician needs/offers information and services related to the devices, procedures and analysis to be performed within the lab structure.…”

“…Note that appropriate fields allow the user to specify parameters like the pathology, the discipline and the user's country (location). A complete and formal definition of the CDT and its usage for data tailoring can be found in [9].…”