Éditeur(s) : HAL CCSD Presses Universitaires de Louvain (distribution I6DOC/CIACO)
Résumé : In French. 10 pages, 5 figures, LaTeX 2e using EPSF and custom package taln2006.sty (designed by Pierre Zweigenbaum, ATALA). Proceedings of the 13th annual French-speaking conference on Natural Language Processing: `Traitement Automatique des Langues Naturelles' (TALN 2006), Louvain (Leuven), Belgium, 10-13 April 2003
National audience
We propose a theoretical framework within which information on the vocabulary of a given corpus can be inferred on the basis of statistical information gathered on that corpus. Inferences can be made on the categories of the words in the vocabulary, and on their syntactical properties within particular languages. Based on the same statistical data, it is possible to build matrices of syntagmatic similarity (bigram transition matrices) or paradigmatic similarity (probability for any pair of words to share common contexts). When clustered with respect to their syntagmatic similarity, words tend to group into sublanguage vocabularies, and when clustered with respect to their paradigmatic similarity, into syntactic or semantic classes. Experiments have explored the first of these two possibilities. Their results are interpreted in the frame of a Markov chain modelling of the corpus' generative processe(s): we show that the results of a spectral analysis of the transition matrix can be interpreted as probability distributions of words within clusters. This method yields a soft clustering of the vocabulary into sublanguages which contribute to the generation of heterogeneous corpora. As an application, we show how multilingual texts can be visually segmented into linguistically homogeneous segments. Our method is specifically useful in the case of related languages which happened to be mixed in corpora.
Verbum ex machina : Actes de la 13ème conférence annuelle sur le Traitement Automatique des Langues Naturelles
Louvain (Leuven), Belgium

hal-00327607
https://hal.archives-ouvertes.fr/hal-00327607
ARXIV : 0810.1212

Éditeur(s) : HAL CCSD IOS Press (Amsterdam)
Résumé : 6 pages, 7 embedded figures, LaTeX 2e using the ecai2006.cls document class and the algorithm2e.sty style file (+ standard packages like epsfig, amsmath, amssymb, amsfonts...). Extends the short version contained in the ECAI 2006 proceedings.
International audience
Without prior knowledge, distinguishing different languages may be a hard task, especially when their borders are permeable. We develop an extension of spectral clustering -- a powerful unsupervised classification toolbox -- that is shown to resolve accurately the task of soft language distinction. At the heart of our approach, we replace the usual hard membership assignment of spectral clustering by a soft, probabilistic assignment, which also presents the advantage to bypass a well-known complexity bottleneck of the method. Furthermore, our approach relies on a novel, convenient construction of a Markov chain out of a corpus. Extensive experiments with a readily available system clearly display the potential of the method, which brings a visually appealing soft distinction of languages that may define altogether a whole corpus.
ECAI 2006: 17th European Conference on Artificial Intelligence
Riva del Garda, Italy

hal-00327782
https://hal.archives-ouvertes.fr/hal-00327782
ARXIV : 0810.1261

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Recently, a large amount of work has been devoted to the study of spectral clustering — a powerful unsupervised classification method. This paper brings contributions to both its foundations, and its applications to text classification. Departing from the mainstream, concerned with hard membership, we study the extension of spectral clustering to soft membership (probabilistic, EM style) assignments. One of its key features is to avoid the complexity gap of hard membership. We apply this theory to a challenging problem, text clustering for languages having permeable borders, via a novel construction of Markov chains from corpora. Experiments with a readily available code clearly display the potential of the method, which brings a visually appealing soft distinction of languages that may define altogether a whole corpus.
ISSN: 0031-3203

hal-00329991
https://hal.archives-ouvertes.fr/hal-00329991
DOI : 10.1016/j.patcog.2008.06.024

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
This work shows the preparation and separation performance assessment of carbon membranes derived from phenolic resin by a vacuum-assisted method and carbonisation in an inert atmosphere. The vacuum time played an important role in tailoring the structure of the membranes. For instance, pore volumes and surface areas increased from 0.81 and 834 to 2.2 cm3g−1and 1910 m2g−1, respectively, as the vacuum time exposure increased from 0 to 1200 s. The significant structural changes correlated very well with water permeation, as fluxes increased by 91% as the vacuum time increased from 0 to 1200 s reaching up to 169 L m−2h−1at 5 bar. Molecular weight cut-off tests showed no rejection for the smaller glucose and sucrose molecules, though this increased to ~ 80% and full rejection for 36 kDa and 400 kDa polyvinyl pyrrolidine. Interestingly, FTIR spectra showed that the peaks of C–H stretching vibration (2800–3200 cm−1) and C–O stretching (1030 cm−1) became more pronounced as a function of increasing vacuum time, strongly suggesting that the use of vacuum further assisted in the polycondensation of phenolic oligomers. Based on these outcomes, a cluster to cluster model is proposed, whereby vacuum application promoted crosslinking reactions of the phenolic resin, creating microporous regions within the clusters, and mesoporous regions between the clusters. © 2016 Elsevier B.V
ISSN: 0376-7388

hal-01683220
https://hal.univ-antilles.fr/hal-01683220
DOI : 10.1016/j.memsci.2016.11.002

Éditeur(s) : HAL CCSD Geological Society of America
Résumé : International audience
In this study, a novel application of a statistical approach is utilized for analysis of downhole logging data from Miocene-aged siliciclastic shelf sediments on the New Jersey Margin (eastern USA). A multivariate iterative nonhierarchical cluster analysis (INCA) of spectral gamma-ray logs from Integrated Ocean Drilling Program (IODP) Expedition 313 enables lithology within this siliciclastic succession to be inferred and, through comparison with the 1311 m of recovered core, a continuous assessment of depositional sequences is constructed. Signifi cant changes in INCA clusters corroborate most key stratigraphic surfaces interpreted from the core, and this result has particular value for surface recognition in intervals of poor core recovery. This analysis contributes to the evaluation of sequence stratigraphic models of large-scale clinoform complexes that predict depositional environments, sediment composition, and stratal geometries in response to sea-level changes. The novel approach of combining statistical analysis with detailed lithostratigraphic and seismic refl ection data sets will be of interest to any scientists working with downhole logs, especially spectral gamma-ray data, and also provides a reference for the strengths and weaknesses of multi component analysis applied to continental margin lithofacies. The method presented here is appropriate for evaluating successions elsewhere and also has value for hydrocarbon exploration where sequence stratigraphy is a fundamental tool.
EISSN: 1553-040X

insu-00858238
https://hal-insu.archives-ouvertes.fr/insu-00858238
DOI : 10.1130/GES00913.1

Éditeur(s) : HAL CCSD
Résumé : International audience
In this paper, we use a multifractal approach based on the computation of two spectrums for image analysis and texture segmentation problems. The two spectrums are the Legendre Spectrum, determined by classical methods, and the Large Deviation Spectrum, determined by kernel density estimation. We propose a way for the fusion of these two spectrums to improve textured image segmentation results. An unsupervised k-means is used as clustering approach for the texture classification. The algorithm is applied on mosaic image built using IKONOS images and various natural textures from the Brodatz album. The segmentation obtained with our approach gives better results than the application of each spectrum separately.
CIARP Proceedings
Cancun, Mexico

hal-00602268
https://hal.archives-ouvertes.fr/hal-00602268

Éditeur(s) : American Geophysical Union
Résumé : Noise with periods 3 to 10 s, ubiquitous in seismic records, is expected to be mostly generated by pairs of ocean wave trains of opposing propagation directions with half the seismic frequency. Here we present the first comprehensive numerical model of microseismic generation by random ocean waves, including ocean wave reflections. Synthetic and observed seismic spectra are well correlated (r > 0.85). On the basis of the model results, noise generation events can be clustered in three broad classes: wind waves with a broad directional spectrum (class I), sea states with a significant contribution of coastal reflections (class II), and the interaction of two independent wave systems (class III). At seismic stations close to western coasts, noise generated by class II sources generally dominates, but it is intermittently outshined by the intense class III sources, limiting the reliability of seismic data as a proxy for storm climates. The modeled seismic noise critically depends on the damping of seismic waves. At some mid-ocean island stations, low seismic damping is necessary to reproduce the observed high level and smoothness of noise time series that result from a spatial integration of sources over thousands of kilometers. In contrast, some coastal stations are only sensitive to noise within a few hundreds of kilometers. This revelation of noise source patterns worldwide provides a wealth of information for seismic studies, wave climate applications, and new constraints on the possible directional distribution of wave energy.
Journal of Geophysical Research ( JGR ) - Oceans (0148-0227) (American Geophysical Union), 2011-09 , Vol. 116 , N. C09004 , P. 21 p.

Droits : 2011 by the American Geophysical Union
http://archimer.ifremer.fr/doc/00044/15531/12926.pdf
DOI:10.1029/2011JC006952
http://archimer.ifremer.fr/doc/00044/15531/

Éditeur(s) : HAL CCSD
Résumé : National audience
Designing systems able to interact with students in a natural manner is a complex and far from solved problem. A key aspect of natural interaction is the ability to understand and appropriately respond to human emotions. This paper details our response to the continuous Audio/Visual Emotion Challenge (AVEC'12) whose goal is to predict four affective signals describing human emotions. The proposed method uses Fourier spectra to extract multi-scale dynamic descriptions of signals characterizing face appearance, head movements and voice. We perform a kernel regression with very few representative samples selected via a supervised weighted-distance-based clustering, that leads to a high generalization power. We also propose a particularly fast regressor-level fusion framework to merge systems based on different modalities. Experiments have proven the efficiency of each key point of the proposed method and our results on challenge data were the highest among 10 international research teams.
IHM'14, 26e conférence francophone sur l'Interaction Homme-Machine
Lille, France

hal-01089628
https://hal.archives-ouvertes.fr/hal-01089628
https://hal.archives-ouvertes.fr/hal-01089628/document
https://hal.archives-ouvertes.fr/hal-01089628/file/p44-nicole.pdf

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
The Blake excursion was among the first recognized with directional and intensity behavior known mainly from marine sediment and Chinese loess. Age estimates for the directional shifts in sediments are poorly constrained to about 118−100 ka, i.e., at the marine isotope stage (MIS) 5e/5d boundary. Moreover, sediments at Lac du Bouchet maar, France and along the Portuguese margin reveal what may be a "post-Blake" excursion at about 105−95 ka. The excursional directions are associated with a prominent paleointensity minimum between about 125 and 95 ka in global stacked records. Lava flow recordings of the Blake excursion(s) have, however, been questionable because precise ages required for correlation with these sediment records are lacking. To establish new, independent records of the Blake excursion, and link these into a larger Quaternary GITS, we have undertaken 40Ar/39Ar incremental heating and unspiked K-Ar experiments on groundmass from the transitionally magnetized Inzolfato flow on Lipari Island. We also obtained 40Ar/39Ar incremental heating results for a lava flow on Amsterdam Island originally thought to record the Mono Lake excursion and from the transitionally magnetized El Calderon basalt flow, New Mexico that was K-Ar dated by Champion et al. (1988) at 128 ± 66 ka. Unspiked K-Ar ages of four samples from the Inzolfato flow are 102.5 ± 4.7, 101.3 ± 3.3, 97.1 ± 2.6, and 96.8 ± 3.1 ka and thus indistinguishable from one another. 40Ar/39Ar results are more complex, with three samples yielding discordant age spectra. Based on incremental heating data obtained in both the UW-Madison and Gif-sur-Yvette 40Ar/39Ar laboratories, a fourth sample yields six concordant age plateaus and a weighted mean age of 105.2 ± 1.4 ka that we take as the best estimate of time since the flow erupted. Five 40Ar/39Ar incremental heating experiments on the Amsterdam Island lava yield a plateau age of 120 ± 12 ka, whereas ages from two sites in the Calderon flow are 112 ± 23 and 101 ± 14 ka, together giving a weighted mean of 104 ± 12 ka. The age of 120 ± 12 ka from Amsterdam Island, though imprecise, correlates with the Blake excursion. In contrast, the 104-105 ka age obtained from both Lipari and New Mexico indicates that these lavas record a younger period of dynamo instability, most probably associated with the post-Blake excursion. These radioisotopic ages are consistent with the astronomical ages of two paleointensity minima in the PISO-1500 global stack. Our findings indicate that the Blake and post-Blake excursions are both global features of past geodynamo behavior and support the hypothesis that Brunhes chron excursions are temporally clustered into two groups of at least a half-dozen each spanning over 220 to 30 ka and 720 to 520 ka.
ISSN: 1871-1014

hal-01054336
https://hal.archives-ouvertes.fr/hal-01054336
DOI : 10.1016/j.quageo.2012.12.005