Haïfa Zargayouna

Alignment-based Partitioning of Large-scale Ontologies

Ontology alignment is an important task for information integration systems that can make different resources, described by various and heterogeneous ontologies, interoperate. However very large ontologies have been built in some domains such as medicine or agronomy and the challenge now lays in scaling up alignment techniques that often perform complex tasks. In this paper, we propose two partitioning methods which have been designed to take the alignment objective into account in the partitioning process as soon as possible. These methods transform the two ontologies to be aligned into two sets of blocks of a limited size. Furthermore, the elements of the two ontologies that might be aligned are grouped in a minimal set of blocks and the comparison is then enacted upon these blocks. Results of experiments performed by the two methods on various pairs of ontologies are promising.

Uncovering the Semantics of Wikipedia Pagelinks

Wikipedia pagelinks, i.e. links between Wikipages, carry an intended semantics: they indicate the existence of a factual relation between the DBpedia entity referenced by the source Wikipage, and the DBpedia entity referenced by the target Wikipage of the link. These relations are represented in DBpedia as occurrences of the generic ”wikiPageWikilink” property. We designed and implemented a novel method to uncover the intended semantics of pagelinks, and to represent them as semantic relations. In this paper, we test our method on a subset of Wikipedia, showing its potential impact for DBpedia enrichment.

YaSemIR: yet another semantic information retrieval system

In this paper we present YaSemIR, a free open-source Semantic Information Retrieval system based on Lucene. It takes one or more ontologies in OWL format and a terminology associated to each ontology in SKOS format to index semantically a text collection. The terminology is used to annotate concepts in documents, while the ontology is used to exploit the taxonomic information in order to expand these with their subsumers. YaSemIR is a flexible system that may be configured to work with different ontologies, on various types of documents.

A Typology of Semantic Relations Dedicated to Scientific Literature Analysis

We propose a method for improving access to scientific literature by analyzing the content of research papers beyond citation links and topic tracking. Our model relies on a typology of explicit semantic relations. These relations are instantiated in the abstract/introduction part of the papers and can be identified automatically using textual data and external ontologies. Preliminary results show a promising precision in unsupervised relationship classification.

Evaluation of Knowledge Based Applications: Benchmark and Guidelines

In this paper we propose to report a work on evaluation of a knowledge based application that leads to the constitution of a benchmark such as those that exist in Information Retrieval evaluations. This benchmark enables to perform quantitative evaluation by classic metrics such as precision and recall. We had also conducted a qualitative analysis that helps the elaboration of guidelines and methodological indications for ontology evaluation and enhancement.

Shallow Parsing of INEX Queries

This article presents the contribution of the LIPN : Laboratoire d’Informatique de Paris Nord (France) to the NLQ2NEXI (Natural Language Queries to NEXI) task (part of the Natural Language Processing (NLP) track) of the Initiative for Evaluation of XML Retrieval (INEX 2006). It discusses the use of shallow parsing methods to analyse natural language queries.

Unsupervised Relation Extraction in Specialized Corpora Using Sequence Mining

This paper deals with the extraction of semantic relations from scientific texts. Pattern-based representations are compared to word embeddings in unsupervised clustering experiments, according to their potential to discover new types of semantic relations and recognize their instances. The results indicate that sequential pattern mining can significantly improve pattern-based representations, even in a completely unsupervised setting.