From prison cells to phd ebook pdf free download

Transcription

1 Loïc MAISONNASSE D.O.B Laboratoire d Informatique de Grenoble 385, rue de la Bibliothèque B.P Grenoble Cedex 9 France Phone office: (+33) Phone mobile: (+33) Ph.D. in Computer Science experience University Pierre Mendes France, LIG laboratory, Grenoble, France, Researcher and Teacher in computer science, IUT2 informatique (computer science department) Three month internship in Singapore at the IPAL laboratory (Image Perception, Access & Language), in collaboration with the NUS (National University of Singapore) University Joseph Fourier, LIG laboratory, Grenoble, France, PhD-student funded by French ministry of research, within the team MRIM (Information retrieval team). I worked on information retrieval. More precisely, I developed, from the language model, a graph-based model for semantic representation, evaluated on IR medical collections University Pierre Mendes France, Grenoble, France, three year contract for teaching, office automation, IUP commerce et vente (business school department) CLIPS-IMAG laboratory, Grenoble, France, development of a user interface for experimentation, 4 months Cabrilog, Grenoble, France, Creation of a Web interface for bug management, 2 months Astrophysics Laboratory of Grenoble, Creation of a calculation software for circumstellar objects detection, 2 months. Education Since 2004 Ph.D. Student in Computer Science, University Joseph Fourier, MRIM team of the LIG laboratory, Grenoble, France, on information retrieval models applied to semantic representation for medical information retrieval My Ph.D. will be defended by mid-may Master in computer science, with honours (1st), University Joseph Fourier, Grenoble, France, master project; A preliminary study on the use of Universal Networking Language in information retrieval 1/5

2 Scientific Activities Interests Natural language processing, Information retrieval, Conceptual indexing, Language modelling. Thesis ( ) Titre Laboratory Supervisors Chair Member Summary Les supports de vocabulaires pour les systèmes de recherche d information orientés précision : application aux graphes pour la recherche d information médicale. Vocabulary supports for precision oriented information retrieval systems: application to graphs for medical information retrieval. MRIM Team, LIG laboratory (Laboratoire Informatique de Grenoble). Pr. Catherine Berrut, university Joseph Fourier, Grenoble, France. Dr. Jean-Pierre Chevallet, LIG, University Pierre Mendès France, Grenoble, France. Pr. Jean-Pierre Giraudin, LIG, University Pierre Mendès France, Grenoble, France. Pr. Patrick Bosc, IRISA, ENSSAT, Lannion, France. Pr. Jean-Marie Pinon, INSA, LIRIS, Lyon, France. In my Ph.D. I explored a framework for the development of precision-oriented information retrieval models. This framework promotes the notion of vocabulary support to model expressive representations used by information retrieval systems. Indeed few modelling framework are available to specify information retrieval systems and we propose such a framework which focuses on the modelling of expressiveness. This framework can be used to choose the expressiveness of a model and to compare models on their level of expressiveness. In this framework we are moving towards the use of an expressive representation of the text. For this, we propose two models that are using representations with strong expressiveness. Both models are based on graphs representations. Through these two models are similar on their expressiveness, they are opposed on their underlying models. Indeed, we implement our first model with a model derived from conceptual graphs, and the second one with a model derived from the language modelling approach to information retrieval. To use these models on text, we propose the use of a two-step process based on language processing that promotes information coverage. The first step produces an intermediate representation of documents in which each sentence is represented by a graph. This step is domain dependent. The second step creates documents final representations from the intermediate one. We finally apply our two models on the medical domain. To do so, we use the meta-thesaurus UMLS 1 and we propose several ways to build the intermediate representation of documents. The effectiveness of our model is proven by a number of experiments on the CLEF medical campaign. This campaign enables us to test our models in a real framework and to compare it to other teams. Indeed while taking part in this campaign in 2007 one of our models got the best results 2. Participation in evaluation campaign Clef 2007 A part of my research activities consists in participating in evaluation campaigns. These campaigns are used to validate some aspects of the research work and to propose new approaches. Participating in such campaigns requires lots of personal investment in program development and in experimentation, but allows for direct comparison with other research teams. In this campaign I have been involved in the medical information task. The purpose of this task is the multilingual search of diagnostics and images. For this task, we (results LIG-MRIM) 2/5

3 Deft 2006 Deft 2005 Clef 2005 proposed to use a conceptual representation and a language model applied to graphs. This approach enabled us to achieve the best results in this task. The purpose of this campaign was the thematic segmentation of documents. For this participation, we demonstrated the usefulness of the combination of different approaches. Our team got the 2 nd best results. This campaign goal was to automatically attribute discourse to its authors. We evaluated different NLP approaches based on lemmatisation and shallow parsing. We obtained the 8 th position on 25 runs. In this campaign we participated to the monolingual and multilingual information retrieval tasks. On these tasks we tested the usability of language models on syntactic structures. Our results were in the average for this campaign.

4

5 Teaching Activities ATER 3 ( ) Department Description IUT2 Computer science department of the University Pierre Mendès France, Grenoble 2, France I have been teaching to an audience consisting of second year students in computer science. I participated in end of year juries, preparation of exams, corrections Courses Audience Hours/year Algorithms and programming Operating System - concurrent programming second year students second year students 48 h tutorial 20 h tutorial 40 practical session Three years contract for teaching ( ) Department Description IUP commerce et vente (business school department), University Pierre Mendes France, Grenoble 2, france I gave courses on office automation for business students. In addition to these courses I participated in student internship evaluation, student recruitment, end of year juries, preparation and correction of exams, exams surveillance, Creation of the course material Courses Word/Excel/ Word/Excel/ Extra-Curricular Activities: People Licence 3rd year Licence grande distribution (Large retailers) Licence Hôtellerie (hotel and catering) Licence bank Participation in doctoriales and entrepreneurship trophies in Grenoble First Aid Certification of the First Aid Training (2006) Sports Reading Roller Hockey, Squash Science-fiction, Fantasy Hours/year Total 44h 20 h 28 h 92 h (22h x 2) (10h x 2) (14h x 2) 28h 17,5 h 24 h (12 h x 2) 8h 45.5 h 24 h 8 h 3 Attaché Temporaire d enseignement et de Recherche. Researcher and Teacher in computer science for one year 5/5