Mohamed Farah | ISAMM - Academia.edu (original) (raw)

Papers by Mohamed Farah

Research paper thumbnail of L'Agr�gation en Recherche d'Information

Coria, 2007

Ce papier donne une nouvelle présentation des modèles standards de la recherche d'information, dé... more Ce papier donne une nouvelle présentation des modèles standards de la recherche d'information, décrits selon deux dimensions. La première porte sur les sources d'évidence qu'utilisent les modèles et la manière dont ils les agrègent pour mesurer l'importance ou le poids d'un terme dans un document. La seconde concerne la manière dont ces poids sont agrégés pour calculer un score de pertinence. Les mécanismes d'agrégation utilisés dans les deux cas sont alors explicités et critiqués motivant le recours à une nouvelle famille de méthodes basées sur de nouveaux mécanismes d'agrégation plus adaptés.

Research paper thumbnail of Novel Approaches in Text Information Retrieval - Experiments in the Web Track of TREC 2004

Text REtrieval Conference, 2004

In this paper, we report our experiments in the mixed query task of the Web track for TREC 2004. ... more In this paper, we report our experiments in the mixed query task of the Web track for TREC 2004. We deal with the problem of ranking Web documents within a mul- ticriteria framework and propose a novel approach for infor- mation retrieval. We focus on the design of a set of crite- ria aiming at capturing complementary aspects of relevance.

Research paper thumbnail of An outranking approach for rank aggregation in information retrieval

Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '07, 2007

Mohamed Farah Lamsade, Paris Dauphine University Place du Mal de Lattre de Tassigny 75775 Paris C... more Mohamed Farah Lamsade, Paris Dauphine University Place du Mal de Lattre de Tassigny 75775 Paris Cedex 16, France farah@lamsade.dauphine.fr ... Daniel Vanderpooten Lamsade, Paris Dauphine University Place du Mal de Lattre de Tassigny 75775 Paris Cedex 16, France ...

Research paper thumbnail of A Multiple Criteria Approach for Information Retrieval

Lecture Notes in Computer Science, 2006

Research in Information Retrieval shows performance improvement when many sources of evidence are... more Research in Information Retrieval shows performance improvement when many sources of evidence are combined to produce a ranking of documents. Most current approaches assess document relevance by computing a single score which aggregates values of some attributes or criteria. We propose a multiple criteria framework using an aggregation mechanism based on decision rules identifying positive and negative reasons for judging

Research paper thumbnail of Novel Approaches in Text Information Retrieval - Experiments in the Web Track of TREC 2004

In this paper, we report our experiments in the mixed query task of the Web track for TREC 2004. ... more In this paper, we report our experiments in the mixed query task of the Web track for TREC 2004. We deal with the problem of ranking Web documents within a mul- ticriteria framework and propose a novel approach for infor- mation retrieval. We focus on the design of a set of crite- ria aiming at capturing complementary aspects of relevance. Moreover, we provide aggregation procedures that are based on decision rules, to get the ranking of relevant documents.

Research paper thumbnail of An outranking approach for information retrieval

Information Retrieval, 2008

Over the last three decades, research in Information Retrieval (IR) shows performance improvement... more Over the last three decades, research in Information Retrieval (IR) shows performance improvement when many sources of evidence are combined to produce a ranking of documents. Most current approaches assess document relevance by computing a single score which aggregates values of some attributes or criteria. They use analytic aggregation operators which either lead to a loss of valuable information, e.g., the min or lexicographic operators, or allow very bad scores on some criteria to be compensated with good ones, e.g., the weighted sum operator. Moreover, all these approaches do not handle imprecision of criterion scores. In this paper, we propose a multiple criteria framework using a new aggregation mechanism based on decision rules identifying positive and negative reasons for judging whether a document should get a better ranking than another. The resulting procedure also handles imprecision in criteria design. Experimental results are reported showing that the suggested method performs better than standard aggregation operators.

Research paper thumbnail of Graphics, Vision and Image Processing Journal, ISSN 1687-398X, Volume 15, Issue 2, Delaware, USA

This issue (DOI: 10.13140/RG.2.1.2640.1365) includes the following articles; P1151517372, G. Srin... more This issue (DOI: 10.13140/RG.2.1.2640.1365) includes the following articles; P1151517372, G. Srinivasa Rao and V.Vijaya Kumar and Suresh Penmesta.Suresh varma, "Cellular Automata Clustering Based on Morphological Reconstruction (CACMR)", P1151536403, P.V.L. Suvarchala and S. Srinivas Kumar and B. Chandra Mohan, "Heterogeneous Features using S-Transform and Local Binary patterns for Non-ideal Iris Recognition ", P1151539407, Kiran Jadhav and Ramesh Kulkarni and Gaurav Tawde, "PERFORMANCE ANALYSIS OF SINGLE DICTIONARY LEARNING FOR SINGLE IMAGE SUPER-RESOLUTION", P1151521383, Rafika Ben Salem and Karim Saheb Ettabaa and Mohamed Ali Hamdi, "Supervised Spectral-Spatial Hyperspectral Image Classification based on Oversampling and Composite Kernels", P1151545425, Hafedh Nefzi and Mohamed Farah and Imed Riadh Farah, "Evaluation of the Taxonomic Consistency of Ontologies based on WordNet Hierarchical and Lexical Relations", P1151547443, Mallikka Rajalingam and Valliappan Raman and Putra Sumari, "An Enhanced Character Segment Method in Image-based Email Classification"

Research paper thumbnail of L'Agr�gation en Recherche d'Information

Coria, 2007

Ce papier donne une nouvelle présentation des modèles standards de la recherche d'information, dé... more Ce papier donne une nouvelle présentation des modèles standards de la recherche d'information, décrits selon deux dimensions. La première porte sur les sources d'évidence qu'utilisent les modèles et la manière dont ils les agrègent pour mesurer l'importance ou le poids d'un terme dans un document. La seconde concerne la manière dont ces poids sont agrégés pour calculer un score de pertinence. Les mécanismes d'agrégation utilisés dans les deux cas sont alors explicités et critiqués motivant le recours à une nouvelle famille de méthodes basées sur de nouveaux mécanismes d'agrégation plus adaptés.

Research paper thumbnail of Novel Approaches in Text Information Retrieval - Experiments in the Web Track of TREC 2004

Text REtrieval Conference, 2004

In this paper, we report our experiments in the mixed query task of the Web track for TREC 2004. ... more In this paper, we report our experiments in the mixed query task of the Web track for TREC 2004. We deal with the problem of ranking Web documents within a mul- ticriteria framework and propose a novel approach for infor- mation retrieval. We focus on the design of a set of crite- ria aiming at capturing complementary aspects of relevance.

Research paper thumbnail of An outranking approach for rank aggregation in information retrieval

Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '07, 2007

Mohamed Farah Lamsade, Paris Dauphine University Place du Mal de Lattre de Tassigny 75775 Paris C... more Mohamed Farah Lamsade, Paris Dauphine University Place du Mal de Lattre de Tassigny 75775 Paris Cedex 16, France farah@lamsade.dauphine.fr ... Daniel Vanderpooten Lamsade, Paris Dauphine University Place du Mal de Lattre de Tassigny 75775 Paris Cedex 16, France ...

Research paper thumbnail of A Multiple Criteria Approach for Information Retrieval

Lecture Notes in Computer Science, 2006

Research in Information Retrieval shows performance improvement when many sources of evidence are... more Research in Information Retrieval shows performance improvement when many sources of evidence are combined to produce a ranking of documents. Most current approaches assess document relevance by computing a single score which aggregates values of some attributes or criteria. We propose a multiple criteria framework using an aggregation mechanism based on decision rules identifying positive and negative reasons for judging

Research paper thumbnail of Novel Approaches in Text Information Retrieval - Experiments in the Web Track of TREC 2004

In this paper, we report our experiments in the mixed query task of the Web track for TREC 2004. ... more In this paper, we report our experiments in the mixed query task of the Web track for TREC 2004. We deal with the problem of ranking Web documents within a mul- ticriteria framework and propose a novel approach for infor- mation retrieval. We focus on the design of a set of crite- ria aiming at capturing complementary aspects of relevance. Moreover, we provide aggregation procedures that are based on decision rules, to get the ranking of relevant documents.

Research paper thumbnail of An outranking approach for information retrieval

Information Retrieval, 2008

Over the last three decades, research in Information Retrieval (IR) shows performance improvement... more Over the last three decades, research in Information Retrieval (IR) shows performance improvement when many sources of evidence are combined to produce a ranking of documents. Most current approaches assess document relevance by computing a single score which aggregates values of some attributes or criteria. They use analytic aggregation operators which either lead to a loss of valuable information, e.g., the min or lexicographic operators, or allow very bad scores on some criteria to be compensated with good ones, e.g., the weighted sum operator. Moreover, all these approaches do not handle imprecision of criterion scores. In this paper, we propose a multiple criteria framework using a new aggregation mechanism based on decision rules identifying positive and negative reasons for judging whether a document should get a better ranking than another. The resulting procedure also handles imprecision in criteria design. Experimental results are reported showing that the suggested method performs better than standard aggregation operators.

Research paper thumbnail of Graphics, Vision and Image Processing Journal, ISSN 1687-398X, Volume 15, Issue 2, Delaware, USA

This issue (DOI: 10.13140/RG.2.1.2640.1365) includes the following articles; P1151517372, G. Srin... more This issue (DOI: 10.13140/RG.2.1.2640.1365) includes the following articles; P1151517372, G. Srinivasa Rao and V.Vijaya Kumar and Suresh Penmesta.Suresh varma, "Cellular Automata Clustering Based on Morphological Reconstruction (CACMR)", P1151536403, P.V.L. Suvarchala and S. Srinivas Kumar and B. Chandra Mohan, "Heterogeneous Features using S-Transform and Local Binary patterns for Non-ideal Iris Recognition ", P1151539407, Kiran Jadhav and Ramesh Kulkarni and Gaurav Tawde, "PERFORMANCE ANALYSIS OF SINGLE DICTIONARY LEARNING FOR SINGLE IMAGE SUPER-RESOLUTION", P1151521383, Rafika Ben Salem and Karim Saheb Ettabaa and Mohamed Ali Hamdi, "Supervised Spectral-Spatial Hyperspectral Image Classification based on Oversampling and Composite Kernels", P1151545425, Hafedh Nefzi and Mohamed Farah and Imed Riadh Farah, "Evaluation of the Taxonomic Consistency of Ontologies based on WordNet Hierarchical and Lexical Relations", P1151547443, Mallikka Rajalingam and Valliappan Raman and Putra Sumari, "An Enhanced Character Segment Method in Image-based Email Classification"