 |
Gian Piero Zarri was awarded a M.Sc. in Electronic Engineering from the
University of Bologna, Italy, and then a Ph.D. in Computer Science from the
University of Paris-Orsay, France. He is actually Research Director at the
French National Centre for Scientific Research (CNRS), and leader of a Data
Engineering team.
Dr. Zarri is known internationally for combining methods from knowledge-based
systems and natural language processing, as well as databases and information
retrieval systems.
In a first phase of his scientific activities (1964-1974) accomplished,
at least partly, at the Centro di Cibernetica e di Attività Linguistiche
of the Milan University (Italy) Dr. Zarri has carried out a pioneering
work in the domain of the use of computers in Literary and Linguistic studies.
He has been one of the first researchers, on a world scale, to apply
computer-based techniques to the study of ancient languages like the Latin
(Roman epigraphy), the Greek (papyrology), or the Meroitic (Meroitic epigraphy:
Meroitic is the language of the ancient Nubians). But his most well-known
activity in this domain has been an original work on the automation of Literary
Criticism i.e., the possibility of reconstructing by computer the
genealogy of medieval manuscripts, with applications both in the Classic and
the Medieval Philology. From a Computer Science point of view, this work can be
reduced, at least partly, to an atypical application of Graph Theory. After
having built up a computer environment, QUENTIN/80, to allow the philologists
to carry on the automatic reconstruction of the genealogies, Dr. Zarri has
organised in 1976, when he already worked at CNRS, the first international
Conference on the subject, see the bibliography. QUENTIN/80 was able, in
particular, to take into account the contamination phenomena,
corresponding to the possibility that a manuscript may be associated with more
than a father-manuscript.
After having been hired as a researcher by CNRS (1974), Mr. Zarri has
reoriented his scientific activities mainly in the fields of Artificial
Intelligence and Knowledge Representation. In this last area, inter alia, he
defined and developed NKRL (short for Narrative Knowledge Representation
Language), the knowledge representation language used in the European
Commission funded projects NOMOS (ESPRIT P5330), in the legal domain, COBALT
(LRE P61-011), in the financial domain, CONCERTO (ESPRIT P29159), in the field
of intelligent indexing and retrieval of information stored on Internet and
textual databases, and EUFORBIA (IAP 2104), in the field of intelligent
filtering of information on Internet. Dr. Zarri was the scientific driving
force behind all these projects; in particular, he is actually (2002) the
scientific co-ordinator of the EUFORBIA project, which involves research,
academic and industrial teams from four European countries (France, Italy,
Portugal, and United Kingdom). NKRL and its predecessors e.g., the
metalanguage developed in the framework of the RESEDA project
(RESEDA = REseau SEmantique DocumentAire, Semantic Network for Information
Retrieval) - have been built up thanks to the financial support of CNRS,
of the European Commission, of INRIA (Institut National de Recherche en
Informatique et Automatique) and of other French Agencies like DGRST
(Délégation Générale à la Recherche Scientifique et
Technique) and ADI (Agence de lInformatique).
More specifically, the aim of NKRL is that of proposing an unambiguous,
language-neutral, standard representation of the meaning of the so-called
narrative documents. In these, the main part of the information
content consists in the description of facts or events which relate the real or
intended behaviour of some actors (characters, personages): these
try to attain a specific result, experience particular situations, manipulate
some (concrete or abstract) materials, deliver or receive messages, etc.
Narrative documents are of a strong industrial and economic interest: they
correspond in fact, inter alia, to news stories, telex reports, corporate
documents (memos, policy statements, reports and minutes), normative texts,
intelligence messages, etc. From a technical point of view, the most important
innovation of NKRL with respect to similar knowledge representation tools
consists in the addition of an ontology of events (a catalogue of standard,
formalised representation of characteristic situations and events) to the usual
ontology of concepts. A new version of the main NKRL tools, written in Java and
XML/RDF compatible, has been realised in the framework of the CONCERTO project.
Dr. Zarris record of research is extensive, and he is the author of more
than one hundred and thirty refereed papers in the fields of Artificial
Intelligence (Knowledge Representation, Expert Systems, Intelligent Databases),
Computational Linguistics, Information Retrieval, Application of Computer
Science Techniques to Humanities and Social Sciences. He has presented his work
in the most prestigious Conferences of both Artificial Intelligence
IJCAI, International Joint Conference on Artificial Intelligence, ISMIS,
International Symposium on the Management of Information Systems, etc.
and Computational Linguistics e.g., COLING, Computational Linguistics
(International) Conference see the bibliography. He is the editor of the
book Operational Expert Systems Applications in Europe (Pergamon
Press, 1991), and co-author, inter alia, of a recent book on Intelligent
Database Systems (Addison-Wesley/ACM Press, January 2001). He received,
in 1977, the bronze medal of CNRS, in recognition of his outstanding scientific
activity.
Dr Zarri is or has been on the editorial board of some important international
scientific journals (Computers and the Humanities, Expert Systems for
Information Management, International Journal of Applied Expert Systems,
International Journal of Failures and Lessons in Information Technology,
International Journal of Expert System: Research and Applications, New Review
of Applied Expert Systems, etc.), and on the Programme Committee of
international conferences on Knowledge-Based Systems, as the World Congress on
Expert Systems, the International Symposium on Methodologies for Intelligent
System (ISMIS), the International Conference on Database and Expert Systems
Application (DEXA), etc..
His present teaching activities concern, mainly, a post-graduate course on
Knowledge Representation at the Institut National de Langues et
Civilisations Orientales (INALCO, a French Grande École) in Paris. Dr.
Zarri is also member of the Computer Science Committee of INALCO.
Dr. Zarri has co-operated intensively with industry. After some years, from
1964 to 1969, spent working at the University of Milan as researcher and
teacher, he was employed from 1969 to 1973 by Praxis SpA, a high-technology
Milanese SME working in the field of process control. More recently, he had the
opportunity to act as consultant both to large industrial groups (e.g.,
Thomson-CSF and Alcatel-Alsthom, France, Olivetti, Italy) and to SMEs (e.g.,
Axon, Portugal, ERLI and STEP Informatique, France, Quinary, Italy). His work
has consisted, mainly, in : a) helping these companies to set up viable
consortia able to submit coherent proposals in answer to EC calls (Esprit, LRE
) ; b) inspiring the technical work in case of acceptance of such
proposals see his role in the context of the NOMOS, COBALT and CONCERTO
projects mentioned before, or of other, previous Esprit projects like ESB
(Expert System Builder, P96) ; c) more in general, executing some
technology transfer activities. In 1985, Dr. Zarri has been on leave at Tecsiel
(Rome, Italy), a firm pertaining to the Finsiel group, as scientific counsellor
of the President and C.E.O. of the company in the initial phase of the
Artificial Intelligence activities of this company. From June 1st, 1986, to
August 31st, 1987, he has been temporarily employed, as project leader in the
Expert Systems field, by CIMSA SINTRA, subsidiary of the Thomson-CSF group. Dr.
Zarri has also been involved as an expert in the R&D activities of the European
Union (information retrieval, Luxembourg) and of UNESCO (expert system
technologies for the underdeveloped countries).
|