Catálogo de publicaciones - libros

In this paper we propose the model of a prototypical NLP architecture of an information access system to support a team of experts in a scientific design task, in a shared and heterogeneous framework. Specifically, we believe AI/NLP can be helpful in several tasks, such as the extraction of implicit information needs enclosed in meeting minutes or other documents, analysis of explicit information needs expressed through Natural Language, processing and indexing of document collections, extraction of required information from documents, modeling of a common knowledge base, and, finally, identification of important concepts through the automatic extraction of terms. In particular, we envisioned this architecture in the specific and practical scenario of the Concurrent Design Facility (CDF) of the European Space Agency (ESA), in the framework of the SHUMI project (Support To HUman Machine Interaction) developed in collaboration with the ESA/ESTEC – ACT (Advanced Concept Team).

Automatic word spacing will be a very useful tool in a SMS (simple message service) , if it can be commercially served. However, the problems of implementing it in the devices such as mobile phones are small memory and low computing power of the devices. To tackle these problems, this paper proposes a combined model of rule-based learning and memory-based learning. According to the experimental results, the model shows higher accuracy than rule-based learning or memory-based learning alone. In addition, the generated rules are so small and simple that the proposed model is appropriate for small memory devices.

When visiting cities as tourists, most users intend to explore the area looking for interesting things to see or for information about places, events, and so on. An adaptive information system, in order to help the user choice, should provide contextual information presentation, information clustering and comparison presentation of objects of potential interest in the area where the user is located. To this aim, we developed a system able to generate personalized presentation of objects of interest, starting from an annotated city-map.

Haptic interfaces may allow blind people to interact naturally and realistically with 3D virtual models of objects that are unsuitable for direct tactile exploration. The haptic interaction can be offered at different scales, by changing the relative size of probe and objects and by organizing different levels of details into the model. In addition, haptic interfaces can actively drive the user along the most effective exploration path around the scene. All these features can significantly help the synthesis and the understanding of the huge amount of tactile sensations (that blinds must collect serially) beyond the limits of the exploration in the real world. The paper describes an architecture (and its already realized modules for visualization, collision detection and force simulation) intended to generate a reliable simulation of the geometrical and physical interactions between the user’s hand and a virtual 3D scene.

We propose an approach to Natural Language Processing exploiting knowledge domain in an e-commerce scenario. Based on such modeling an NLP parser is presented, aimed at translating demand/supply advertisements into structured Description Logic expressions, automatically mapping sentences with concept expressions related to a reference ontology.

In this paper, an adaptive natural language dialog model for Web-based cooperative interactions is proposed to improve the results in achieving a successful filtered search on the Web. The underlying principle, based on automatically generating language-driven interactions which take into account the context and the user’s feedback is discussed. The preliminary working design and experiments, and the results of some real evaluations are also highlighted.

A model-based approach for minimization of test sets for human-computer systems is in tro duced. Test cases are efficiently generated and se lected to cover both the behavioral model and the complementary fault model of the system under test (SUT). Results known from state-based conformance testing and graph theory are used and extended to construct algorithms for minimizing the test sets.

The present work investigates the problem of determining a learning path inside a suitable domain ontology. The proposed approach enables the user of a web learning application to interact with the system using natural language in order to browse the ontology itself. The course related knowledge is arranged as a three level hierarchy: content level, symbolic level, and conceptual level bridging the previous ones. The implementation of the ontological, the interaction, and the presentation component inside the TutorJ system is explained, and the first results are presented.

It is proposed a step towards the automatic description of scenes with a geometric approach. The scenes considered are composed by a set of elements that can be geometric forms or iconic representation of objects. Every icon is characterized by a set of attributes like shape, colour, position, orientation. Each scene is related to a set of sentences describing its content. The proposed approach builds a data driven vector semantic space where the scenes and the sentences are mapped. Sentences and scene with the same meaning are mapped in near vectors and distance criteria allow retrieving semantic relations.

This paper proposes man-machine interface of support system for analyzing answers to open-ended questions supplied by customers of the mobile game content reply when they unsubscribe from the services. Since open-ended questions, however, place no restrictions on descriptions, the answers include an enormous amount of text data for the content provider. It is time-consuming to read all of the texts one by one. Since a large number of answers are identical to choices included in the multiple-choice questions or unnecessary opinions unconcerned with the game, there are few answers that should be read. Most opinions are needed to know only the number and the outline. However, the provider should not omit to read the unexpected opinion that is a minority. Additionally, since answers are input through cellular phones, they often include many symbols dependent on various kinds of terminals and grammatical mistakes, making them hard to understand. Our research, therefore, aims to create a system that supports the provider to analyze the answers of open-ended questions efficiently. The main function of the support system divides the answers into typical opinions and atypical opinions. Divided opinions are presented with different user interfaces, because the content providers can analyze the two type opinions with each way.