Khaled Khelif

Ontology-Based Semantic Annotations for Biochip Domain

We propose a semi-automatic method using information extraction (IE) techniques for generating ontology-based annotations for scientific articles, useful for biologists working in the biochip domain.

Recognising Professional-Activity Groups and Web Usage Mining for Web Browsing Personalisation

Web usage mining can play an important role in supporting the navigation on the future Web. In fact detection of common or professional profiles allows browsers and web sites to personalise the user session and to recommend specific resources to the interested people. Semantic web approach seems interesting for this task. We propose in this paper a generic approach for profile detection relying on semantic web technologies. It takes advantages from ontologies, semantic annotations on web resources and inference engines.

Semantic web technologies for interpreting DNA microarray analyses: the MEAT system

This paper describes MEAT (Memory of Experiments for the Analysis of Transcriptomes), a project aiming at supporting biologists working on DNA microarrays. We provide methodological and software support to build an experiment memory for this domain. Our approach, based on Semantic Web Technologies, is relying on formalized ontologies and semantic annotations of scientific articles and other knowledge sources. It can probably be extended to other massive analyses of biological events (as provided by proteomics, metabolomics...).

Semantic Patent Clustering for Biomedical Communities

In this paper, we present the Pat Clust clustering solution for textual documents based on semantic criteria. Our proposition is dedicated to patent documents of the biomedical domain. We present three different approaches and we show that semantic web techniques clearly allow to improve the quality of resulting clusters.

Using the Intension of Classes and Properties Definition in Ontologies for Word Sense Disambiguation

We present an ontology-driven word sense disambiguation process. The main idea consists of using the context of the ambiguous word to decide which class can be assigned to it. The disambiguation relies on similarities between classes assigned to the ambiguous word, classes assigned to terms close to it in the text, and on the type of properties that could occur between them. The computation of the similarity uses domain ontologies to provide semantic distances based on definitions in intension. We tested our approach in the extraction of annotations from biomedical texts.

Semi-automatic Construction of an Ontology and of Semantic Annotations from a Discussion Forum of a Community of Practice

In this paper we describe a method for creating a semantic portal from a corpus of e-mails of a community of practice. Using Natural Language Processing (NLP) techniques to build semi-automatically an ontology and semantic annotations from an email-list corpus raises several original issues. The ontology and the annotations thus obtained are then used in a semantic portal that facilitates ontology-guided and personalized navigation of the CoP members.

MEAT: an experiment memory for interpreting DNA microarray analyses

This paper describes MEAT (Memory of Experiments for the Analysis of Transcriptomes), a project aiming at supporting biologists working on DNA microarrays. We provide methodological and software support to build an experiment memory for this domain. Our approach, based on Semantic Web Technologies, is relying on formalized ontologies and semantic annotations of scientific articles and other knowledge sources. It can probably be extended to other massive analyses of biological events.