Catálogo de publicaciones - libros

This paper addresses the problem of automatic learning of scenarios. A ubiquitous computing environment must have the ability to perceive its occupants and their activities in order to recognize a context and to provide appropriate services. A context (a scenario) can be modeled as a temporal sequence of situations. Hard coding contexts by hand is a complex task. Our goal is to learn these context models based on a set of videos showing actors playing predefined scenarios. Once these models are learned, we can use them to classify new scenarios. Hidden Markov Models (HMMs) are particularly well suited for problems with a strong temporal structure; they are easily adaptable to variability of input and robust to noise. But two problems need to be addressed: how many HMMs do we need for all possible scenarios and how many states for each HMM. We propose in this paper an approach based on an incremental algorithm addressing these two problems. Under the best conditions we obtained the minimal error rate of 1.96% (2 errors in 102 validation entries).

The paper presents the architecture and design of a description logics (DLs) reasoner based on the Model Driven Architecture (MDA) methodology. The architecture relies on OMG’s description logics metamodel, tableau metamodel, and model transformations using a language for model transformation. We show an example of DLs knowledge base using UML notation in context of MDA. The idea can be easily applied to implementation of a first-order logic theorem prover. The paper does not discuss implementation details of reasoning algorithms and the reasoner performance.

Guideline formalization is recognized as an important component in improving computerized guidelines, which in turn leads to better informedness, lower inter-practician variability and, ultimately, to higher quality healthcare. By means of a modeling exercise, we investigate the role of guideline formalization tools which use two different knowledge transformation principles in producing re-usable knowledge objects useful for representing medical processes and performing updates of medical guidelines. We give a general evaluation of usefulness and state the main requirements for tools that reuse medical knowledge and support authoring of guidelines.

We present an original navigation approach to explore data in the framework of the semantic web. Before navigation, data are represented and integrated by jointly using Topic Maps and description logics. Our navigation approach improves the traditional web navigation with two specificities. First, the navigation paths are semantic instead of structural links. Second, it is a subject-centric instead nothing-centric. These two facilities increase the efficiency of information retrieval in the Web. They are implemented in an adaptive interface, which browses, gradually, data as a concept map with rest to user navigation.

AdviceWeb is a delivery environment aimed at enabling a wide range of high-quality rule-based advisory expert systems to be created and made available free of charge to users via the world wide web. Applications have a uniform ‘look and feel’, do not require any special software to use and are fully integrated with other sources of information on the web. A rapid application development facility is provided which automatically converts modules written in a simple scripting language into rules and questions for a production rule interpreter running on the AdviceWeb server.

Recently, several standards have emerged for ontology markup languages that can be used to formalize all kinds of knowledge. However, there are no widely accepted standards yet that define APIs to manage ontological data. Processing ontological information still suffers from the heterogeneity imposed by the plethora of available ontology management systems. Moreover, ubiquitous computing environments usually comprise software components written in a variety of different programming languages, which makes it even more difficult to establish a common ontology management API with programming language agnostic semantics. We implemented an ontological Knowledge Base Server, which can expose the functionality of arbitrary off-the-shelf ontology management systems via a formally specified and well defined API. A case study was carried out in order to demonstrate the feasibility of our approach to use an ontological Knowledge Base Server as a registry for ubiquitous computing systems.

Although on-line shops have become very popular lately, there are still some sectors of industry, such as clothing and furniture, where the on-line shop concept has not penetrated the market yet. The main reason for that is the diversity of the specific sectors, the mentality of the market players and the lack of user friendly environments, where the customer could actually “see” the products. This paper presents an Intelligent Configurable Electronic Shop Platform, based on Ontologies and 3D Visualization, that aims at enabling a suitable representation of products with the most realistic possible visualization outcome. The platform, designed for the furniture sector, includes all the practicable electronic commerce variants and its on-line product configuration process is controlled by an ontology that was created using the OWL Web Ontology Language.

User interfaces are commonly designed to match the requirements of the so-called ‘typical user’ who represents an abstract generalization of each user of an application. In systems employed by different kinds of users (such as Web systems), the identification of the requirements of their users is very difficult, if not impossible. Our proposal is to avoid the construction of interactive dialogs during the design stage, building them dynamically once the specific cognitive, perceptual and motor requirements of the current user of the system are known: that is, during the execution stage.

Logic reasoning presents notable difficulties for young children. This paper presents Logic Model Creator (LMC), a new environment that supports building and exploration of intuitive visual representation of logic models by young children. LMC logic models are structured as hypothesis, decision and / or counter decision components. These models are built using visual entities which represent the learning concepts of a specific domain. In this paper we focus on the architecture of LMC and the basic functionality of the environment. In particular we describe the the dynamic creation of equivalent logic models according to the so-called Reference Logic Model, constructed by the students’ tutor. Furthermore an assessment module which provides immediate advice to the student in order to help them create a valid logic model is presented. Through experimentation it is demonstrated that the users of LMC can have rich interaction and assessment while exploring decision making logic constructs.

This paper presents a novel approach to detect interaction groups in intelligent environments. To understand human activity, we must identify human actors as well as their interpersonal links. Interaction detection is a good cue to address activity of user groups. An attentional model is derived from gravitational model and cognitive psychology approaches. Whereas determining locally users’ focus of attention is a difficult task, this model exploits contextual elements such as position, speed and saliency of objects in the scene to estimate shared attention. The attentional model shows promising results on simulated scenarios where unexpected events occur.