Catálogo de publicaciones - libros

Web services are becoming the basis for electronic commerce of all forms. The number of services being provided is increasing but different service providers use different ontologies for services’ descriptions. This has made it difficult for service discovery agents to compare and locate the desired services. Inputs and outputs are important pieces of information that can be used when searching for the needed services. Therefore, in this paper, to facilitate users or software agents for discovering Web services in multi-ontology environments, we propose an approach to determine the semantic similarity of services’ inputs/outputs that are described by different ontologies.

The adoption of a policy-based approach for the dynamic regulation of a system or service (e.g. security, QoS or mobility service) requires an appropriate policy representation and processing. In the context of the Semantic Web, the representation power of languages enriched with semantics (i.e. semantic languages), together with the availability of suitable interpreters, make such kind of languages well suited for policies representation. In this paper, we describe our proposal for the combination of the CIM-OWL ontology (i.e., the mapping of the DMTF Common Information Model into OWL) with the Semantic Web Rule Language as the basis for a semantically-rich security policy language that can be used to formally describe the desired security behaviour of a system or service. An example of security policy in this language and its reasonig are also presented.

In this paper we identify trust relationships among users and systems. We try to adhere to simplicity principle in our modelling of the system. By using simple model and free lightweight technologies, we show that it is possible to implement secure Web applications/services. The paper also addresses some security problems and issues about implementing Web Services.

UDDI is not suitable for handling semantic markups for Web services due to its flat data model and limited search capabilities. In this paper, we introduce an approach to allow for support of semantic service descriptions and queries using registries that conforms to UDDI V3 specification. Specifically, we discuss how to store complex semantic markups in the UDDI data model and use that information to perform semantic query processing. Our approach does not require any modification to the existing UDDI registries. The add-on modules reside only on clients who wish to take advantage of semantic capabilities. This approach is completely backward compatible and can integrate seamlessly into existing infrastructure.

Demands for web mapping services are increasing worldwide since the rise of Internet which became a growing medium to disseminate geospatial information. However, these services need to be more reliable, accessible and personalized. They also need to be improved in terms of data format, interoperability and on-the-fly processing and transfer. In this paper we propose a multiagent-based approach to generate maps on-the-fly in the context of web mapping applications. Our approach is abls to adapt the contents of maps in real-time to users’ needs and display terminals. It also speeds up map generation and transfer. Our approach which is called , combines different techniques: multiagent systems, cartographic generalization and multiple representations.

In this paper we concentrate on conceptual modeling and semantics of service delegation and alignment in information systems. In delegation, a source company wishes to hand over parts of its functionality together with related responsibilities to a supplying party. From the side of the outsourcer the search for a suitable supplier mostly will be a manual process with all the consequences of a long time to market, as well as trial and error before a good fit is obtained between both related parties. This paper addresses an agent-based solution for improving this match-making process in B2B markets. Part of the match-making process will be the alignment of business processes on the side of the outsourcer as well on the side of the supplier. We will provide a formal means to ensure that the delegation relationship, determined by a ruling service level agreement (SLA), satisfies specific correctness criteria. These correctness criteria are defined in terms of consistency and completeness between the delegated operation and the associated operation offered by the supplier. Our correctness criterion will concern mappings between an existing delegator schema and an existing supplier schema, and will address both semantical and ontological aspects pertaining to delegation and alignment. Agent-based delegation together with formal specifications can prove their value in the process of constructing delegation contracts. Our analysis will be performed within the modeling framework based on the UML/OCL formalism. The concepts we discussed in this paper are illustrated by an example of companies delegating billing services to Billing Service Providers.

Context awareness is increasingly forming one of the key strategies for delivering effective information services in mobile contexts. The limited screen displays of many mobile devices mean that content must be carefully selected to match the user’s needs and expectations, and context provides one powerful means of performing such tailoring. Context aware mobile systems will almost certainly become ubiquitous – already in the United Kingdom affordable ’smartphones’ include GPS location support. With this hardware comes the opportunity for ’on-board’ applications to use location data to provide new services – until recently such systems could only be created with complex and expensive components. Furthermore, the current ’mode’ of the phone (e.g. silent, meeting, outdoors), contents of the built-in calendar, etc. can all used to provide a rich context for the user’s immediate environment.

The immaturity of the field of context-aware computing means that little is known about how to incorporate appropriate personalisation mechanisms into context-aware applications. One of the main challenges is how to elicit and represent complex, context-dependent requirements, and then use the resulting representations within context-aware applications to support decision-making processes. In this paper, we characterise several approaches to personalisation of context-aware applications and introduce our research on personalisation using a novel preference model.

Context-awareness is a fundamental aspect of the ubiquitous computing paradigm. In this framework, a relevant problem that has received little attention is the large heterogeneity of formats used to express a context information: text files in ad-hoc format, HTTP header, XML files over specific DTD’s, RDF, CC/PP and so on. So many applications meet difficulties in interpreting and integrating context information coming from different sources. In this paper we propose an approach to this problem. We first present a general architecture for context-aware adaptation that is able to take into account different coordinates of adaptation. We then show how, in this framework, external profiles are dynamically captured and translated into a uniform common representation that is used by the system to meet the requirements of adaptation. We also present a prototype application that implements the proposed approach.

Recently, there has been a significant increase in the use of data via the mobile web. Since the user interfaces for mobile devices are inconvenient for browsing through many pages and searching their contents, many studies have focused on ways to recommend content or menus that users prefer. However, the mobile usage pattern of content or services differs according to context. In this paper, we apply context information—location, time, identity, activity, and device—to recommend services or content on the mobile web. A Korean mobile service provider has implemented context-aware recommendations. The usage logs of this service are analyzed to show the performance of context-aware recommendations.