Damir Jurić

Building OWL ontologies by analyzing relational database schema concepts and WordNet semantic relations

Ontologies represents main building material for creation of semantic Web. Without ontologies it could not be possible to enable machine processability of data on the Semantic Web. Data is usually stored inside relational databases that represent the backbone for dynamical web sites. That data represents the invisible Web, ignored by search engines and Web crawlers. This paper presents a way to extract semantics from relational database schema and create OWL ontology automatically. Because relational database schema can capture very limited semantics, ontologies are enriched with additional WordNet semantics.

Uncovering the Deep Web: Transferring Relational Database Content and Metadata to OWL Ontologies

Organizing the publicly available Web content into highly systematized domain ontologies is a necessary step in the evolvement of the Semantic Web. A large portion of that content called the deep Web is stored in relational databases and it is not accessible to Web search engines. Incorporation of the deep Web data results in domain ontologies richer both in content and in semantic relations. In this paper we introduce a framework for an automatic mapping of relational database metadata and content to domain ontologies written in OWL. Relational constructs: relations, attributes and primary-foreign key associations are translated to OWL classes, datatype properties and object properties. Database tuples become ontology instances. In order to define reference points for integration with other ontologies the constructed ontologies are further enriched with additional semantics from the WordNet lexical database using word sense disambiguation mechanisms. A software implementation of the approach has been developed and evaluated on case study examples.

Distributed architecture for association rule mining

Organizations have adopted various data mining techniques to support their decision-making and business processes. However, the mining analysis is not performed and supervised by the final user, the management of the organization, since the knowledge of mathematical models as well as expert database administration skills is required. This paper describes a distributed architecture for association rule mining analysis in the retail area, designed to be used directly by the management of an organization and implemented as a Java web application. The rule discovery algorithm is executed at the database server that hosts the source data warehouse, while the only used client tool is a web browser. The user interactively initiates the rule discovery process through a simple user interface, which is used later to browse, sort and compare the discovered rules.

Improving search on WWW.HR web directory by introducing ontologies

In this paper we propose ontology-based improvement of the Croatian Web directory search mechanism, which is currently not capable of executing queries that take into account the structure and semantics of users query. The proposed approach is verified by introducing an ontology in the domain (directory category) of “tourism”. We address three problems related to the search mechanism: low recall, high recall and low precision and vocabulary mismatch. The results show significant improvements in terms of subjective relevance and quality of results.

Informativeness of inflective noun bigrams in croatian

A feature of Croatian and other Slavic languages is a rich inflection system, which does not exist in English and other languages that traditionally dominate the scientific focus of computational linguistics. In this paper we present the results of the experiments conducted on the corpus of the Croatian online spellchecker Hascheck, which point to using non-nominative cases for discovering collocations between two nouns, specifically the first name and the family name of a person. We analyzed the frequencies and conditional probabilities of the morphemes corresponding to Croatian cases and quantified the level of attraction between two words using the normalized pointwise mutual information measure. Two components of a personal name are more likely to co-occur in any of the non-nominative cases than in nominative. Furthermore, given a component of a personal name, the conditional probability that it is accompanied with the other component of the name are higher for the genitive/accusative and instrumental case than for nominative.

Learning semantic n-ary relations from Wikipedia

Automated construction of ontologies from text corpora, which saves both time and human effort, is a principal condition for realizing the idea of the Semantic Web. However, the recently proposed automated techniques are still limited in the scope of context that can be captured. Moreover, the source corpora generally lack the consensus of ontology users regarding the understanding and interpretation of ontology concepts. In this paper we introduce an unsupervised method for learning domain n-ary relations from Wikipedia articles, thus harvesting the consensus reached by the largest world community engaged in collecting and classifying knowledge. Providing ontologies with n-ary relations instead of the standard binary relations built on the subject-verb-object paradigm results in preserving the initial context of time, space, cause, reason or quantity that otherwise would be lost irreversibly. Our preliminary experiments with a prototype software tool show highly satisfactory results when extracting ternary and quaternary relations, as well as the traditional binary ones.