1,721,021 research outputs found
EVALITA 2007: The Named Entity Recognition Task
No full textIn this paper we describe the Named Entity Recognition Task organized in the context of the EVALITA 2007 evaluation campaign. In particular, we report information about the dataset and the evaluation metrics we used, and we discuss the results obtained by participant systems
A Plug-in Approach for an Integrated Consultation of Generic and Specialized Wordnets
No full textAlthough generic (i.e. domain independent) and specialized (i.e. domain specific) lexical resources are usually developed with different aims, an integrated consultation of the two resources would be useful for many practical purposes. We describe a plug-in approach which takes into account our intuitions about an integrated consultation. The model is based on the definition of plug-in relations that are established to consider possible overlapping and inconsistencies between the two resources. As a consequence, the inheritance of linguistic oriented information makes the specialized resources usable in existing wordnet based applications. The approach has been experimented connecting ItalWordNet, a generic lexical database, and Economic-WordNet, a specialized wordnet for the economic and financial domai
A Photograph of the Gender Distribution in the Research Personnel at ITC-irst in 2004. What changes have occurred since 1990?
No full textThis report illustrates some statistics related to research personnel and gender at ITC-irst, the Center for Scientific and Technological research of ITC - Istituto Trentino di Cultura, Trento (Italy). After a short description of the organization of the institute and some methodological notes, the report presents the gender-disaggregated data of ITC-irst research personnel for the year 2004, separated per contract level, department and contract duration; changes occurred during the period 1990-2004 have also been highlighted, again on the basis of gender-disaggregated data. As a result of the analysis of the data, it has been found that in terms of absolute numbers, the presence of women researchers at ITC-irst has been considerably increasing since 1990. Indeed, at the end of 1990, women researchers at ITC-irst numbered only 13 (11%), reaching 65 (22%) by the end of 2004. On the other hand, the analysis shows that the situation has not changed with respect to contract levels, as the distribution of men and women still follows a clearly scissors-shaped trend. Finally, the analysis of the data for departments raises a question: are cultural and social aspects the only causes of the weak position of women inside the research field or could a gender-sensitive management positively influence the situation
Merging Global and Specialized Linguistic Ontologies
No full textThere is an increasing interest in linguistic ontologies (e.g. WordNet) for a variety of content-based tasks, including conceptual indexing, word sense disambiguation and cross-language information retrieval. A relevant contribution in this direction is represented by linguistic ontologies with domain specific coverage, which are a crucial topic for the development of concrete application systems. This paper tries to go a step further in the direction of the interoperability of specialized linguistic ontologies, by addressing the problem of their integration with global ontologies. This scenario poses some simplifications with respect to the general problem of merging ontologies, since it enables to define a strong precedence criterion so that terminological information overshadows generic information whenever conflicts arise. We assume the EuroWordNet model and propose a methodology to `plug` specialized linguistic ontologies into global ontologies. Experimental data related to an implemented algorithm, which has been tested on a global and a specialized linguistic ontology for the Italian language, are provide
Assessing Answer Quality in Task-oriented Question Answering
No full textIn this paper we present a method to evaluate Question Answering systems in restricted domain applications, which present peculiar features with respect to open domain Question Answering. While the answers provided by the system are still evaluated manually as in QA evaluation based on Information Retrieval, we propose to assess the overall quality of answers rather than concentrating merely on answers’ correctness. We propose to evaluate each answer from different perspectives, i.e. not only from the point of view of the exactness of the answer, but of the data provided as well, while at the same time taking into consideration aspects connected to comprehensibility and answer presentation
Integrating Generic and Specialized Wordnets
No full textAlthough generic (i.e. domain independent) and specialized (i.e. domain specific) lexical resources are usually developed with different aims, an integrated consultation seems to be necessary for many NLP based applications. We describe an integration procedure based on the definition of plug-in relations that are established to manage overlaps and inconsistencies between the two resources. The approach has been experimented connecting ItalWordNet, a generic lexical database for italian, and Economic-WordNet, a specialized wordnet for the economic and financial domai
Semantic Coordination for Document Retrieval
No full textWe present CtxMatch, an algorithm that finds mappings between two heterogeneous partially overlapping Classification Hierarchies (e.g. taxonomic structures used to organize documents). CtxMatch relies on the semantic interpretation of both the labels provides in the CHs and the hierarchical structures of the Chs; it does not consider the content of classified documents, thus allowing the retrieval of any kind of documents (e.g. text files, images, applications, videos, etc.). The Web Directories of Google and Yahoo! have been chosen as an evaluation set for discussing the performance of CtxMatc
Making Hidden Semantics of Hierarchical Classifications Explicit
No full textConcept hierarchies are semi-structured knowledge repositories used for organizing large amounts of documents. File systems, products taxonomies for the market place and the directories provided by Web portals are common examples of concept hierarchies. We take the perspective in which such knowledge sources are inherently distributed and we address the problem of allowing their interoperability. In this paper first we provide a formal semantics for concept hierarchies and then we use that formal framework to explore a number of linguistic issues crucial for interpreting the implicit knowledge represented there. Relevant phenomena addressed include word sense disambiguation WORDNET(r) has been used as sense repository), the explicitation of multiwords' semantics and the interpretation of coordinations. The Web directories of Google and Yahoo has been considered for a number of case studie
Towards Interactive Question Answering: An Ontology-Based Approach
No full textThe ability to provide both rich and natural answers with respect to a given question, and clear explanations for failures, is a crucial aspect for a future generation of Question Answering systems able to interact with a user. We argue that such abilities are necessarily based on a deep analysis of the content of both the question and the answer, and propose an ontology-based approach to represent the structure of a question-answer pair in the context of utterance. The approach is domain and language-independent and can be considered a general framework for both Open Domain Question Answering and Natural Language Interfaces to Databases. We provide definitions for an Interactive Question Answer (IQA) ontology which capture significant aspects of interaction, and suggest how dialogue templates based on the IQA ontology can model natural and rich dialogues with a user
- …
