This paper presents the general objectives of the ONTOTEXT project (From Text to Knowledge for the Semantic Web), and the activities carried out during the first year of its development cycle. First, the task of annotating huge amounts of textual data (e.g. those available on the Web or in local document collections) will be introduced, focusing on its importance in order to enhance the interoperability of such data through ontology-based reasoning. Then, the main issues related to the annotation task will be discussed. These include the choice of an adequate formalism to capture and describe different types of relevant information contained in a text, and the adaptation of existing languagespecific markup formalisms to a new language (Italian in our case). Finally, the results of our experience in the concrete annotation of information about people and temporal expressions for the Italian Content Annotation Bank (I-CAB) being developed at ITC-irst and CELCT will be reported.
[PDF document]

We 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 provided 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 CtxMatch.

Classification Hierarchies (CHs) are widely used to organize documents in a way that makes their retrieval easier. Common examples of CHs are Web directories, marketplace catalogs, and file systems. In this paper we discuss and evaluate \ctxmatch, an approach to interoperability that discovers mappings among CHs considering the semantic interpretation of their nodes. \ctxmatch\ performs a linguistic processing of the labels attached to the nodes, including tokenization, Part of Speech tagging, multiword recognition and word sense disambiguation. We present an evaluation of the overall performance of the approach over Web directories as well as a systematic analysis of the linguistic modules involved.
[PDF document]

This paper describes the main characteristics of the ItalWordNet semantic databese, built in the context of the SI-TAL Italian NationalProject, within which a set of integrated resources and tools for the automatic treatment of the Italian language was realized. The database was created by extending the Italian wordnet developed within the EuroWordNet project, by adding: i) adjectives, adverbs and proper nouns (not dealt with in EuroWordNet); ii) a terminological subset related to the economic-financial domain. The relevant changes involved by these extensions both in the linguistic model and in the data structure are also illustrated. In particular we we discuss: i) the overall architecture of the database; ii) the semantic relations used to encode information on synsets; iii) the changes made to the EuroWordNet Top Ontology structure; iv) the specific characteristics of the terminological subset and the solutions adopted to link it to the generic wordnet.

Most of the data stored in the Semantic Web is organized in schema models, which can be represented as labeled graphs where labels are short natural language expressions. Examples of schema models include ER-schema automata, ontologies, taxonomies, and Web Directories. The semantics of schema models is not explicit but is hidden in their structures and labels. To obtain semantic interoperability we need to make their semantics explicit by taking into account both the interpretation of the labels and the structures described by the arcs. We propose a methodology for interpreting schema models on the basis of the taxonomic relations and the linguistic material they contain. We rely on a set of linguistic repositories, such as WordNet, and explore a number of crucial linguistic issues such as disambiguation of polysemous words, multiwords, and coordinations. The Web Directories of Google and Yahoo! have been chosen as an evaluation set. We show that there is a considerable amount of information to be made explicit and discuss the performance of an implementation of our analysis.
[PS document]

Hierarchical classifications are concept hierarchies used to organize large amounts of documents. File systems, products' taxonomies for the market place and the directories provided by Web portals are common examples of hierarchical classifications. We propose a methodology for building a semantic interpretation of hierarchical classifications on the basis of the analysis of the taxonomic relations and the linguistic material they contain. We provide a formal semantics for hierarchical classifications and use it to interpret the implicit knowledge represented. Relevant phenomena addressed include the disambiguation of polysemous words, the semantics of multiwords, and the interpretation of coordinations. We report about experiments performed on the Web Directories of Google and Yahoo!.
[PS document]

This paper describes an automatic algorithm of meaning negotiation that enables semantic interoperability between local overlapping and heterogeneous ontologies. Rather than reconciling differences between heterogeneous ontologies, this algorithm searches for mappings between concepts of different ontologies. The algorithm is composed of three main steps: (i) computing the linguistic meaning of the label occurring in the ontologies via natural language processing, (ii) contextualization of such a linguistic meaning by considering the context, i.e. the ontologies, where a label occurs; (iii) comparing contextualized linguistic meaning of two ontologies in in order to find a possible matching between them.
[PS document]

This paper describes an automatic algorithm of meaning negotiation that enables semantic interoperability between local overlapping and heterogeneous ontologies. Rather than reconciling differences between heterogeneous ontologies, this algorithm searches for mappings between concepts of different ontologies. The algorithm is composed of three main steps: (i) computing the linguistic meaning of the label occurring in the ontologies via natural language processing, (ii) contextualization of such a linguistic meaning by considering the context, i.e. the ontologies, where a label occurs; (iii) comparing contextualized linguistic meaning of two ontologies in order to find a possible matching between them.
[PS document]