Дисертації з теми "Détection du stress"
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-48 дисертацій для дослідження на тему "Détection du stress".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте дисертації для різних дисциплін та оформлюйте правильно вашу бібліографію.
Soury, Mariette. "Détection multimodale du stress pour la conception de logiciels de remédiation." Thesis, Paris 11, 2014. http://www.theses.fr/2014PA112278/document.
This thesis focuses on the automatic recognition of human stress during stress-inducing interactions (public speaking, job interview and serious games), using audio and visual cues.In order to build automatic stress recognition models, we used audio cues computed from subjects' voice captured via a lapel microphone, and visual cues computed either form subjects' facial expressions captured via a webcam, or subjects' posture captured via a Kinect. Part of this work is dedicated to the study of information fusion form those various modalities.Stress expression and coping are influenced both by interpersonal differences (personality traits, past experiences, cultural background) and contextual differences (type of stressor, situation's stakes). We evaluated stress in various populations in data corpora collected during this thesis: social phobics in anxiety-inducing situations in interaction with a machine and with humans; apathologic subjects in a mock job interview; and apathologic subjects interaction with a computer and with the humanoid robot Nao. Inter-individual and inter-corpora comparisons highlight the variability of stress expression.A possible application of this work could be the elaboration of therapeutic software to learn stress coping strategies, particularly for social phobics.Key words: stress, social phobia, multimodal stress detection, stress audio cues, stress facial cues, stress postural cues, multimodal fusion
Lebègue, Nathalie. "Radicaux libres dérivés de l'oxygène et antioxydants : méthodes de détection, approche biologique et environnementale." Paris 5, 1999. http://www.theses.fr/1999PA05P109.
Gotti, Guillaume. "Modification de surfaces électrochimiques par des nanoparticules d'or pour la détection de molécules impliquées dans le stress oxydant." Toulouse 3, 2013. http://thesesups.ups-tlse.fr/2229/.
Oxidative stress is a biological phenomenon resulting from an imbalance between oxidant and antioxidant species which can be involved in the early stages of many pathologies such as cancer or neurodegenerative diseases. The detection of molecules involved in the oxidative stress is a major issue in terms of public health. Due to its simplicity, fast-response time and low cost, electrochemistry is a method of interest for the detection of dissolved oxygen and hydrogen peroxide, two precursors of more deleterious reactive species in a biological medium. In this work a study of the electrochemical behavior of O2 and H2O2 under neutral conditions was performed by cyclic and steady-state voltammetries, first on bulk unmodified materials (glassy carbon and gold). In order to be further used as a reference and fill a lack in the literature, the charge transfer coefficients were determined for O2 reduction and H2O2 oxidation using Tafel and Koutecky-Levich methods and were compared to those obtained in acidic and basic media. In a second part, gold nanoparticles functionalized glassy carbon electrodes were prepared by two ways: direct electrodeposition from a gold precursor and deposition by adsorption of colloidal gold solution synthesized in aqueous medium. To demonstrate the electrocatalytic properties of those gold nanoparticles, the charge transfer coefficients were determined for O2 reduction and H2O2 oxidation and compared with those obtained on bulk materials. From the different materials used, separate calibrations were made for the two target molecules, and then a simultaneous detection was proposed by cyclic voltammetry to achieve the proof of concept for future sensor
Kotchi, Serge Olivier. "Détection du stress hydrique par thermographie infrarouge : application à la culture de la pomme de terre." Thesis, Université Laval, 2004. http://www.theses.ulaval.ca/2004/22198/22198.pdf.
Water stress affect the yield and the crop quality. For more than twenty years, remote sensing allowed several tools developments to assess water stress detection using sensors and model such as Crop Water Stress Index (CWSI). Sparse crop can generate error in the interpretation of this index. The use of an imaging-radiometer combined to a high spatial resolution visible image help to remove this weakness. This study present a new water stress index based in the use of infrared thermography. A first dataset was taken from greenhouse experimental set up on potatoes species (Highlite and Chieftain) using an hand-held imaging-radiometer (ThermaCAM SC 2000). Repeated measurements were made for water-stress induced plants and for well irrigated plants at different growing stages. The study showed that an early detection of water deficit by infrared thermography is possible as well as the detection of the significant response of plant to heat stress and leaf area change based on water stress intensity. Temperature differences between crop canopy and air (Tc - Ta) are strongly correlated with the water stress. The study has permitted the development of a new index named Area Water Stress Index (AWSI) and the results obtained with this index for water stress detection are very encouraging.
Ladeira, costa claudio Nuno filipe. "GADD34 : Lien moléculaire entre la détection des pathogènes et les voies intégrées de réponses au stress." Thesis, Aix-Marseille, 2012. http://www.theses.fr/2012AIXM4025/document.
Dendritic cells (DCs) are the most important antigen presenting cells. In response to inflammatory stimulation, DCs display a distinct pattern of differentiation that exhibits specific mechanisms to control the immune response. In this work the responses to dsRNA were analyzed. We have shown that in response to a mimic of dsRNA, polyriboinosinic:polyribocytidylic acid (poly I:C), DCs mount a specific integrated stress response during which the transcription factor ATF4 and the growth arrest and DNA damage-inducible protein 34 (GADD34), a phosphatase 1 (PP1) cofactor, are expressed. GADD34 is important to counteract phosphorylation of eIF2α by PKR. In contrast to murine embryonic fibroblasts (MEFs), DCs resist to protein synthesis inhibition induced in response to cytosolic dsRNA. Nevertheless, GADD34 expression does not have a major impact on global protein synthesis. Importantly, GADD34 was shown to be absolutely required for type I-IFN and IL-6 production by MEFs and DCs in response to dsRNA. This observation has important implications in linking pathogen detection with the integrated stress response pathways. The importance of this link is further underlined by the extreme susceptibility of GADD34-deficient fibroblasts and neonate mice to Chikungunya virus infection
Meunier, Anne. "Méthodes analytiques pour la détection de phénomènes biologiques de sécrétion à l'échelle de la cellule unique." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2011. http://tel.archives-ouvertes.fr/tel-00858915.
Lavarde, Marc. "Fiabilité des semi-conducteurs, tests accélérés, sélection de modèles définis par morceaux et détection de sur-stress." Paris 11, 2007. http://www.theses.fr/2007PA112266.
This thesis deals with the using of accelerating data and regression model selection for high technology field: semiconductor chips. The accelerating trail gives us regression frameworks. The aim of the accelerating test consists on fitting the logarithm of the lifetime through the use of some function f, called the acceleration function. However, accelerating data may have misleading and complex comportment. In order to adapt the model with such data, we have proposed to detect the changes on the comportment of the acceleration function. We have considered a collection of piecewise acceleration models candidate to the estimation. For each model candidate we have estimated the least-squares estimation. And we have selected the final estimator using a penalized criterion. The penalized estimator is optimal approximation of the reality since the quadratic risk of penalized estimator is bounded by the minimal risk upon every least-squares estimators candidates. Moreover, this oracle inequality is non asymptotic. Furthermore, we have considered classical reliability cases: the Lognormal case associating with some fatigue failure, and the Weibull case associating with some choc failure. Lastly we have implemented model selection tools in order to realise survey study without a priori on the acceleration models and to use overstress trials
Ghandour, Hala. "Détection par microélectrodes de flux attomolaires d'espèces réactives de l'oxygène et de l'azote produites par un macrophage." Paris 6, 2006. http://www.theses.fr/2006PA066037.
Delaunay-Moisan, Agnès. "Contrôle de l'homéostasie redox et détection des oxydants : Régulation du facteur de transcription Yap1 chez S. cerevisiae." Paris, Institut national d'agronomie de Paris Grignon, 2002. http://www.theses.fr/2002INAP0013.
Gréard, Camille. "La détection des variants alléliques comme voie d'amélioration génétique des plantes fourragères : exemple de la luzerne." Thesis, Poitiers, 2019. http://www.theses.fr/2019POIT2261.
Lucerne (Medicago sativa) is an autotetraploid forage legume, whose breeding could beneficiate from allele mining. This strategy is based on the natural diversity and consists in seeking alleles with a potential effect on the phenotype. The interest of this approach was evaluated by studying five genes of agronomic interest: CAD1 and CCoaOMT (digestibility), CONSTANS-like (forage yield), NHX1 (salt tolerance) and WXP1 (drought tolerance). The diversity of these five genes was evaluated by sequencing 387 genotypes of cultivated accessions and 20 genotypes of wild accessions. The results confirmed a bottleneck during lucerne domestication and selection. CONSTANS-like and WXP1 were very variable whereas CAD1, CCoaOMT and NHX1 contained very few variants. Variants with a potential strong impact on the phenotype were identified in conserved parts of protein sequence within the Faboideae. The impact on phenotype was studied for two mutations of the CONSTANS-like gene: constans-634, causing a premature stop codon and constans-4111, located in a conserved region of the gene. Genotypes carrying one to three doses of the mutations (AAAB, AABB and ABBB) were polycrossed in order to obtain offsprings with every allele combination (AAAA, AAAB, AABB, ABBB and BBBB). KASPar markers were developed to determine the mutation doses in offspring progeny. No homozygous genotype was found for constans-634 in the 1505 offspring progeny. This mutation induced a premature flowering of three days for the genotypes carrying three doses of the mutation. The mutation constans-4111 induced an additive effect on stem height and the homozygous genotypes without the variant where on average 11.8 cm shorter than homozygous genotypes carrying three or four doses of the variant. The application of allele mining strategy in plant schemes of heterozygous autotetraploid species was discussed
Ni, Zhuoya. "Méthode pour l'estimation de la fluorescence de la chlorophylle et son application pour la détection précoce du stress hydrique." Thesis, Strasbourg, 2016. http://www.theses.fr/2016STRAD022/document.
Sun-induced chlorophyll fluorescence is a new way to monitor the vegetation change and global carbon cycle. Through the model simulated analysis, the pot experiment and the airborne flying experiment, the research on detecting the multi-scale sun-induced chlorophyll fluorescence is developed in this dissertation. The main conclusions and innovations are as follows: 1. The maize water control experiments demonstrate that the passive fluorescence can be used to detect the crop water stress, and the analysis of the different responses of the fluorescence and temperature illustrates that the fluorescence is much sensitive to the early water stress. 2. Analyze the effects of temperature, sun zenith angle and fluorescence quantum efficiency on the qualitative fluorescence retrieval, and propose a qualitative fluorescence retrieval method based on the reflectance index. 3. Analyze the effects of airborne fluorescence retrieval, and obtain that sun zenith angle and airborne sensor height are the important factors to affect the sun-induced fluorescence retrieval from the simulated analysis and airborne flying experiment
Cazale, Arnaud. "Développement de microcapteurs chimiques intégrés pour la détection de l'ion sodium en phase liquide: application au suivi du stress physiologique." Phd thesis, Université Paul Sabatier - Toulouse III, 2012. http://tel.archives-ouvertes.fr/tel-00782580.
Cazalé, Arnaud. "Développement de microcapteurs chimiques intégrés pour la détection de l'ion sodium en phase liquide : application au suivi du stress physiologique." Toulouse 3, 2012. http://thesesups.ups-tlse.fr/2449/.
During emergency operations, the first responders (firefighters, civil defense, police or military operation) are subject to significant physiological constraints responsible for serious medical accidents (dehydration, heat stroke. . . ). In the current state, it is absolutely impossible to get a realistic physiological state emergency personnel via health monitoring systems. The work done in this thesis has been to improve the prevention of first responders in developing an integrated headband physiological biochemical sensors able to analyze and monitor the concentration of sodium in sweat. Experiments have demonstrated a physiological link between body condition and hyperthermic sodium ion concentration in sweat. The integration of technology pNa-ISFET in a physiological band, as well as results from in vivo measurements performed on a panel of ten individuals are presented in this manuscript
Depayras, Ségolène. "Etude des mécanismes de détection, d'adaptation et de protection d'une souche de Pseudomonas fluorescens isolée de l'air en réponse au NO2 gazeux, marqueur de pollution automobile." Thesis, Normandie, 2019. http://www.theses.fr/2019NORMR005.
Nitrogen oxides (NOx) atmospheric pollutants, mainly constituted of NO, NO2 and derived compounds, are a big threat to the environment and health. Their chemical properties are largely exploited at the cellular scale for their role in diverse physiological processes such as signalisation (nervous and cardiovascular systems) or in pathogens eradication (immunity system).However, dysregulation in production pathways or exogenous input of these compounds lead to several pathologies (e.g. respiratory diseases), usually attributed to atmospheric pollution. However, a wide range of airborne microorganisms are constantly exposed to these deleterious compounds, intimately connected to reactive oxygen species (ROS). Thus, the hypothesis of this work deals with the impact of NO2, the main atmospheric NOx, on an airborne P. fluorescens, a strain usually neglected but yet associated with human airways, and potentially pathogenic. Following an exposure to 45 ppm of NO2, the survival of P. fluorescens MFAF76a is severely impaired, suggesting a bacteriostatic effect, as comforted by NO2 impact on energetic metabolism. Moreover, an exposure to NO2 induces an envelope stress through the loss of an Unknown Glycerophospholipid (UGP) and the reorganisation of membrane constituents (LPS, peptidoglycan, fatty acids). The efflux pump MexEF-OprN is involved in membrane stabilization and could also efflux NOx, as highlighted by a MFAF76a-oprN mutant. The major porin OprF could also contribute in external membrane stabilisation, however its implication is still under investigation. Moreover, ROS and NOx are interconnected as illustrated by their shared signalisation (OxyR, IscR) and detoxification pathways. The flavohemoprotein Hmp is a crucial element in the detoxification of NOx in P. fluorescens as illustrated in an MFAF76a-hmp mutant. The similarities between the known effects of NO and those observed in the case of an exposure to NO2, suggest a non-enzymatic conversion of NO2, following cell penetration, into NO. Henceforth, deeper studies are required to decode (i) the mechanisms involved in the regulation of the RND efflux pump MexEF-OprN and the flavohemoprotein Hmp, (ii) other relevant actor implicated in the envelope stress response and in detoxification pathways as well as (iii) the fate of NO2 within the cell
Erard, Marie. "Détection par ultramicroélectrodes de flux femtomolaires émis par une cellule vivante isolée : application au stress oxydatif et à l'exocytose de neurotransmetteurs." Paris 6, 2002. http://www.theses.fr/2002PA066124.
Delaunay, Agnès. "Contrôle de l'homéostasie redox et détection des oxydants. Régulation du facteur de transcription Yap1 chez S. cerevisiae." Phd thesis, INAPG (AgroParisTech), 2002. http://tel.archives-ouvertes.fr/tel-00005660.
Daum-Badouard, Carine. "Les lésions des acides nucléiques : détection par CLHP-SM/SM dans les milieux biologiques humains et intérêt comme biomarqueurs du stress oxydant et de l'inflammation." Université Joseph Fourier (Grenoble), 2006. http://tel.archives-ouvertes.fr/tel-00134563.
Molecules involved in oxidative stress or inflammation can damage DNA or RNA, the biopolymers that contain the genetic information. The cel! has developed several enzymatic systems to repair the damage but sorne of them can remain and lead to mutagenecity. Hence, we focused our attention on the simultaneous quantification of severallesions using high performance liquid chromatography coupled to tandem mass spectrometry, a very sensitive, fast and reliable analytical system. The aim of this work was to develop new biomarkers of oxidative stress among the known DNA les ions by quantifying them in human biological fluids. For such a purpose, three types of DNA lesions were monitored : les ions due to oxidative stress, chlorinated lesions arising from inflammation processes and lesions generated subsequently to lipid peroxidation. The analytical method has been optimisel and the biological validation was performed using different pathologies such as diabetes, patients suffered from cancer and treated by radiotherapy and men with infertility. Preliminary results obtained show different levels of DNA lesions between patients and healthy controls. These results have to be confirmed in order to confirm the significance of these lesions as potential biomarkers. Other kinds of Datholoaies will be investigated usina the same approach
Mandon, Céline. "Mise au point d'un bioessai à cellules entières, pour la détection de pollutions, basé sur la technologie du promoteur de stress." Lyon 1, 2005. http://www.theses.fr/2005LYO10276.
Peigné, Cassie-Marie. "Modalités fines d'activation antigénique des LT Vγ9Vδ2 humains : mécanismes de détection du stress cellulaire et implication de la butyrophiline BTN3A/CD277". Nantes, 2016. https://archive.bu.univ-nantes.fr/pollux/show/show?id=f2d45559-cfa5-42ae-a547-6aaf17d1ae3c.
Vγ9Vδ2 T cells are the major sub-population of γδ T cells in human blood. They play a leading part in protection of the organism against infectious agents and tumor cells. Vγ9Vδ2 T cells are specifically and strongly activated by small organic pyrophosphate molecules termed phosphoantigens (PAg) in a TCR dependant way. PAg are metabolites from the isoprenoid pathway shared by both procaryotes and eucaryotes. The activation modalities of Vγ9Vδ2 T cells by PAg remain to be defined in detail. Results from our team revealed an important role played by the BTN3 (CD277) molecule during Vγ9Vδ2 T cell antigenic activation. BTN3, and especially the BTN3A1 isoform, is involved in activating Vγ9Vδ2 T cells. The work described in this thesis, aimed at investigating the underlying mechnisms allowing BTN3 molecule to activate Vγ9Vδ2 T cells. We showed that the B30. 2 intracellular domain of BTN3A1 was able to fix PAg with a histidine residue in position 351 playing a vital role. Furthermore, we identified other proteins that could interact with this intracellular domain. We used the yeast double-hybrid technique, and identified six putative partner proteins. Finally, we carried out a function maping of the intracellular part of the various BTN3 isoforms. These results brought new insights into the antigenic activation modalities of Vγ9Vδ2 T cells by PAg and clarify the key role played by the BTN3 molecule in this process
Loayza, Loza Hildo. "Suivi expérimental du rendement de fluorescence des couverts végétaux par des techniques actives et passives. Application à la détection du stress hydrique." Electronic Thesis or Diss., Sorbonne université, 2022. http://www.theses.fr/2022SORUS465.
The chlorophyll fluorescence (ChlF) is directly related to the photosynthetic process. However, at canopy level this physiological link between fluorescence and photosynthesis may be blurred by structural vegetation changes and geometrical effects linked to interactions between sunlight and the three-dimensional structure of the canopy. Furthermore, much of our knowledge about the relationship between fluorescence and the physiological status of plants come from leaf level studies carried out under laboratory conditions. The physiological significance of ChlF at canopy level and under natural conditions is still a major subject of research and a source of uncertainties in the interpretation of SIF. This doctoral project aims were: 1. To study chlorophyll fluorescence yield at canopy level: we describe a new instrument, Ledflex, which is a micro-LIDAR dedicated to perform continuous measurements of vegetation fluorescence yield. Ledflex has been successfully applied under full sunlight conditions to establish the signature of water-stress on a pea (Pisum Sativum) canopy. Under well-watered conditions the Fs diurnal cycle present an M shape with a minimum (Fmin) at noon which is higher than the fluorescence level observed at predawn (Fo). After several days withholding watering, Fs decreases and Fmin
Guerrier, Sterenn. "High bandwidth detection of mechanical stress in optical fibre using coherent detection of Rayleigh scattering." Electronic Thesis or Diss., Institut polytechnique de Paris, 2022. http://www.theses.fr/2022IPPAT004.
Telecommunication fibres are being deployed all over the world, connecting distant people, institutions, companies with an outstanding quality of service in terms of data rate and latency. Their strategic value in terms of global economy and daily life is now undeniable. Monitoring such an infrastructure has become mandatory, and that far beyond the standard case of breaks localization. From a broader standpoint, optical fibres are an alternative to electro-dynamic point sensors, with a strong asset: the capability to detect and localize multiple independent phenomena all along a fibre. Thus, the millions of kilometers of currently deployed optical fibre around the world constitute a huge potential base of sensors. Distributed vibration sensors have a huge potential regarding sensing of dynamic events, detecting of multiple acoustic signatures up to speech signals. In this thesis, we show how distributed optical fibre sensors can be designed on top of telecommunication fibres, namely standard single mode fibres, and we explore their potential in terms of reach, detection threshold, and sensing bandwidth. We present the interrogator systems for distributed fibre sensing and build a dual-polarization numerical model of such an interrogation system and fibre sensor. Secondly, we tackle the coherent fading issue by means of frequency diversity in the digital domain, i.e. directly applicable at the modulation of the interrogation sequences, before entering the optical domain. We developed MIMO-OFDM which retrieves independent channel estimations from a single fibre segment; the estimations are further combined, and the obtained estimations are assessed with regards to the reliability metric. Throughout this thesis, many experimental measurements were conducted, assessing the capabilities of the Coherent-MIMO interrogator on single-mode-fibre sensors in terms of reach, bandwidth, and detection threshold. We also demonstrate the co-propagation of a sensing signal along with high data rate channels, without any impact on the transmitted data, paving the way to the enhanced monitoring and telemetry in deployed telecommunication networks
Wagner, Nicolas. "Détection des modifications de l’organisation circadienne des activités des animaux en relation avec des états pré-pathologiques, un stress, ou un événement de reproduction." Thesis, Université Clermont Auvergne (2017-2020), 2020. http://www.theses.fr/2020CLFAC032.
Precision livestock farming consists of recording parameters on the animals or their environment using various sensors. In this thesis, the aim is to monitor the behaviour of dairy cows via a real-time localisation system. The data are collected in a sequence of values at regular intervals, a so-called time series. The problems associated with the use of sensors are the large amount of data generated and the quality of this data. The Machine Learning (ML) helps to alleviate this problem. The aim of this thesis is to detect abnormal cow behaviour. The working hypothesis, supported by the biological literature, is that the circadian rhythm of a cow's activity changes if it goes from a normal state to a state of disease, stress or a specific physiological stage (oestrus, farrowing) at a very early stage. The detection of a behavioural anomaly would allow decisions to be taken more quickly in breeding. To do this, there are Time Series Classification (TSC) tools. The problem with behavioural data is that the so-called normal behavioural pattern of the cow varies from cow to cow, day to day, farm to farm, season to season, and so on. Finding a common normal pattern to all cows is therefore impossible. However, most TSC tools rely on learning a global model to define whether a given behaviour is close to this model or not. This thesis is structured around two major contributions. The first one is the development of a new TSC method: FBAT. It is based on Fourier transforms to identify a pattern of activity over 24 hours and compare it to another consecutive 24-hour period, in order to overcome the problem of the lack of a common pattern in a normal cow. The second contribution is the use of fuzzy labels. Indeed, around the days considered abnormal, it is possible to define an uncertain area where the cow would be in an intermediate state. We show that fuzzy logic improves results when labels are uncertain and we introduce a fuzzy variant of FBAT: F-FBAT
Rosé, Elmacin Roseline. "Génétique de la tolérance à la chaleur chez le porc : caractérisation de la variabilité génétique en milieu tropical humide." Thesis, Antilles, 2017. http://www.theses.fr/2017ANTI0170/document.
The aim of the thesis is to characterize the genetic variability of heat tolerance in pork P.ngrowth.Initially, the effect of two climatic events (temperate, TEMP vs. tropical moist, TROP) on production performance and thermoregulatory responses of growing pigs was evaluated.In a second step, we characterized the genetic determinism of heat adaptation in growing pigs.The two analyzes made it possible to propose regions that significantly affect: growth characteristics, ingestion, feed efficiency, blubber thickness and thermoregulatory response characteristics on the chromosome on SSC 2, 5, 8, 10, 11, and 15. Mutations in the MC4R and IGF2 genes appear to have an effect on body temperatures. Interactions between these mutations and ions on the enome have been detected
Tlemsani, Fatima Zohra. "Mesure des transferts thermiques et hydriques par intégration des fluxmètres thermiques textiles dans un vêtement pour les enfants en situation de polyhandicap." Electronic Thesis or Diss., Université de Lille (2022-....), 2023. http://www.theses.fr/2023ULILN004.
Children with cerebral palsy experience significant psychological stress during rehabilitation. This is related to many psychological factors such as fear, anxiety and phobias, and physical factors such as the weight of the rehabilitation devices, their friction on the body, and the pain related to motor problems. In the state of art, it has been shown that researchers have followed an approach using physiological parameters as biomarkers of stress. They mainly use biosignals such as skin temperature (ST), electrocardiography (ECG), electrodermal activity (EDA), electromyography (EMG), respiration, pupil diameter, electroencephalography (EEG) for stress assessment. Since thermal and hydric exchanges are a function of temperature evolution, they can also be an indicator of stress, especially since they represent an indicator of thermal discomfort. For this purpose, in this work, a textile heat fluxmeter, which has characteristics of permeability, flexibility and suitability for use on the skin, has been developed, analyzed and characterized. An experimental device was set up in order to establish a calibration system of the fluxmeter. Then the thermo-hydric behavior of the fluxmeters was analyzed under laboratory conditions. The developed textile heat fluxmeter showed similar sensitivities as the gold standard sensor. Moreover, the study of the fluxmeter performance showed a similar behavior to that of the standard sensor. Therefore, stress tests were conducted on 20 healthy adult volunteers of different ages and genders, women and men, and on two children, 7 and 12 years old, also healthy. Three different types of activities were performed to induce stress, namely, mathematical activities, virtual reality games and a sports activity. This was with the objective of stimulating different types of stress, i.e. positive stress (eustress), negative stress and physical stress, respectively. The results of the tests show a similar behavior between the two fluxmeters (textile and standard), and a positive correlation between the behavior of the electrocardiogram and the fluxmeter. A relation was established in the majority of cases between the volunteers' feedback on the stress they felt and their thermo-hydric response measured by the textile heat fluxmeter
Guénette, Caroline. "Évaluation du potentiel de l’infrarouge spectral pour la détection hâtive du stress nutritif chez les végétaux de grandes cultures. Application à la culture de la pomme de terre." Thesis, Université Laval, 2003. http://www.theses.ulaval.ca/2003/21037/21037.pdf.
At the present time, the management of fertilizers in the agricultural productions is a subject on which producers and administrators grant much attention. The overuse of fertilizers generates important environmental and financial costs which recently lead to both federal and provincial legislations in this field. To gain optimal use of nitrogenized fertilizers, one must evaluate correctly the plant‘s deficiencies. This research project aims at evaluating the potential of detection of the nutritive nitrogen deficiency on potato plants by the use of infra-red spectral remote sensing. To achieve our goal, we performed repeated measurements in a greenhouse experimental setup. Four levels of nitrogen deficiencies were induced on potato plants. The stress caused by the nutritive deficiency was measured according to agronomic techniques and remote sensing techniques, including the infra-red spectroscopy. Once the potential of detection evaluated for each technique, the comparison among them made it possible to determine which technique(s) allowed the hastiest discrimination during the growing season. An optimal combination of index has been identified. Its capacity of discrimination between levels of nitrogen deficiency has been quantified by a discriminant analysis. Research concludes with a notable improvement of discrimination potential by combining three remote sensing index.
Guénette, Caroline. "Évaluation du potentiel de l'infrarouge spectral pour la détection hâtive du stress nutritif chez les végétaux de grandes cultures. Application à la culture de la pomme de terre." Master's thesis, Université Laval, 2003. http://hdl.handle.net/20.500.11794/17814.
At the present time, the management of fertilizers in the agricultural productions is a subject on which producers and administrators grant much attention. The overuse of fertilizers generates important environmental and financial costs which recently lead to both federal and provincial legislations in this field. To gain optimal use of nitrogenized fertilizers, one must evaluate correctly the plant‘s deficiencies. This research project aims at evaluating the potential of detection of the nutritive nitrogen deficiency on potato plants by the use of infra-red spectral remote sensing. To achieve our goal, we performed repeated measurements in a greenhouse experimental setup. Four levels of nitrogen deficiencies were induced on potato plants. The stress caused by the nutritive deficiency was measured according to agronomic techniques and remote sensing techniques, including the infra-red spectroscopy. Once the potential of detection evaluated for each technique, the comparison among them made it possible to determine which technique(s) allowed the hastiest discrimination during the growing season. An optimal combination of index has been identified. Its capacity of discrimination between levels of nitrogen deficiency has been quantified by a discriminant analysis. Research concludes with a notable improvement of discrimination potential by combining three remote sensing index.
Rovere, Martina. "Étude fonctionnelle de la famille des facteurs de transcription ERF-VIIs chez Medicago truncatula : régulateurs clés de l’adaptation au manque d’oxygène." Thesis, Université Côte d'Azur (ComUE), 2018. http://www.theses.fr/2018AZUR4037/document.
Legume crops are known for their capacities to establish a symbiotic relationship with nitrogen fixing soil bacteria. This mutualism culminates in the formation of a new plant organ, the root nodule, in which the symbiont converts atmospheric nitrogen (N2) into ammonia, which can be directly consumed by plants. In nodules, bacterial nitrogenase enzyme is inhibited by traces of oxygen (O2) so different mechanisms maintain this organ at low O2 level. At the same time, nodules need to maintain a high ATP level to support the nitrogenase activity, which is highly energy demanding. Thus, a balance between a tight protection from O2 and an efficient energy production, referred as the “O2 paradox” of N2-fixing legume nodules, has to be reached. In Arabidopsis thaliana, a direct oxygen sensing mechanism has recently been discovered involving members of the ethylene responsive factors (ERFs) group VII. These transcription factors (TFs) possess a characteristic N-terminal amino acid with a cysteine residue at the second position that, under normal O2 conditions, leads to protein degradation following a specific pathway called the N-end rule pathway. Furthermore, it was shown that both O2 and nitric oxide (NO) are required to destabilize the ERFs VII and that a reduction in the availability of either gas is sufficient to stabilize these proteins. Therefore, the goal of this thesis was to investigated the role of ERF-VII family in O2 sensing and adaptation to hypoxia in M. truncatula, model plant for legumes, and to understand how NO interacts with O2 in hypoxic signalization in the microoxic environment that characterizes the nodule. We identified four genes belonging to the ERF-VII TF family in the M. truncatula genome, which present a strong similarity with ERF-VII of Arabidopsis. The characterization of this family at the transcriptional level revealed that only MtERF-B2.2 is up-regulated by hypoxia stress and during nodule development. The three others, MtERF-B1.1, MtERF-B1.11 and MtERF-B2.3 are found constitutively expressed in leaves, roots and nodules. To investigated the protein stability of MtERF-B2.1, the closest orthologous to AtRAP2.12 described as O2-sensors in Arabidopsis, in function of O2/NO availability, we realized a fusion protein with the luciferase reporter protein. Our results on Arabidopsis protoplasts indicated that the N-terminal part of MtERF-B2.1 drives its O2-dependent degradation by the N-end rule pathway. The function of MtERF-B2.1 and MtERF-B2.11 was also investigated both in response to hypoxia stress and during the nodulation process using an RNA interference strategy. Silencing of MtERFB2.1 and MtERF-2.11 showed a significant lower activation of several core hypoxia-responsive genes such as ADH1, PDC1, nsHb1 and AlaAT. These double knock-down transgenic roots were also affected in symbiotic interaction with a significant reduction of the nodulation capacity and nitrogen fixation activity in mature nodules. Overall, the results reveal that O2 sensing mechanism is mediated by ERF-VIIs in M. truncatula roots and nodules and that this mechanism, together with downstream targets, is involved in the organ development and ability to efficiently fix nitrogen. Furthermore, results indicated that MtERF-B2.1/B2.11 are positive regulator of the anaerobic metabolism and the Hb-NO cycle– related genes likely in order to activate alternative ATP generation pathways
Braik, William. "Détection d'évènements complexes dans les flux d'évènements massifs." Thesis, Bordeaux, 2017. http://www.theses.fr/2017BORD0596/document.
Pattern detection over streams of events is gaining more and more attention, especially in the field of eCommerce. Our industrial partner Cdiscount, which is one of the largest eCommerce companies in France, aims to use pattern detection for real-time customer behavior analysis. The main challenges to consider are efficiency and scalability, as the detection of customer behaviors must be achieved within a few seconds, while millions of unique customers visit the website every day,thus producing a large event stream. In this thesis, we present Auros, a system for large-scale an defficient pattern detection for eCommerce. It relies on a domain-specific language to define behavior patterns. Patterns are then compiled into deterministic finite automata, which are run on a BigData streaming platform. Our evaluation shows that our approach is efficient and scalable, and fits the requirements of Cdiscount
Albakour, Subhy. "Stream-automl : automated machine learning overimbalanced data streams for bipartite ranking problems." Electronic Thesis or Diss., Institut polytechnique de Paris, 2024. http://www.theses.fr/2024IPPAT015.
Despite its popularity in the scientific literature, stream learning has yet to substantiate its practical utility in industrial applications. Characterized by the incessant influx of high-velocity, voluminous, and dynamically changing data, online marketing seems to be the favorite candidate for stream learning to make its entry into the industry. In this context, state-of-theart stream learning is of little utility, as it mainly focuses on classification, while bipartite ranking constitutes better modeling of the problem of online marketing. Recently, the combination of stream learning and AutoML, i.e., Stream-AutoML, has been drawing more attention from the scientific community. This work investigates the applicability of Stream-AutoML to bipartite ranking problems when data is imbalanced. We commence by developing a framework to execute and evaluate Stream-AutoML pipelines of stream learning models. Then we propose a framework for computing AUC-ROC incrementally, as well as introducing exponential decay to serve as a forgetting mechanism. We also propose a framework for concept drift detection using AUC-ROC, for which we develop six statistical tests for differences in AUC-ROC with theoretical bounds of type I and type II errors. Finally, we propose four data generators that enrich the tool kit to evaluate concept drift detectors under controlled environments. Results have shown that the proposed methods reduce the resources allocated for evaluation considerably and detect concept drifts with very small false positives. These contributions prepare the field for Stream-AutoML to solve bipartite ranking problems, which can be then exploited in online marketing applications. Optimized implementations of the proposed methods were developed and have already been adopted in the online marketing product of IDAaaS
Bellas, Anastasios. "Détection d'anomalies à la volée dans des signaux vibratoires." Thesis, Paris 1, 2014. http://www.theses.fr/2014PA010020.
The subject of this Thesis is to study anomaly detection in high-dimensional data streams with a specific application to aircraft engine Health Monitoring. In this work, we consider the problem of anomaly detection as an unsupervised learning problem. Modern data, especially those is-sued from industrial systems, are often streams of high-dimensional data samples, since multiple measurements can be taken at a high frequency and at a possibly infinite time horizon. More-over, data can contain anomalies (malfunctions, failures) of the system being monitored. Most existing unsupervised learning methods cannot handle data which possess these features. We first introduce an offline subspace clustering algorithm for high-dimensional data based on the expectation-maximization (EM) algorithm, which is also robust to anomalies through the use of the trimming technique. We then address the problem of online clustering of high-dimensional data streams by developing an online inference algorithm for the popular mixture of probabilistic principal component analyzers (MPPCA) model. We show the efficiency of both methods on synthetic and real datasets, including aircraft engine data with anomalies. Finally, we develop a comprehensive application for the aircraft engine Health Monitoring domain, which aims at detecting anomalies in aircraft engine data in a dynamic manner and introduces novel anomaly detection visualization techniques based on Self-Organizing Maps. Detection results are presented and anomaly identification is also discussed
Baudin, Alexis. "Cliques statiques et temporelles : algorithmes d'énumération et de détection de communautés." Electronic Thesis or Diss., Sorbonne université, 2023. http://www.theses.fr/2023SORUS609.
Graphs are mathematical objects used to model interactions or connections between entities of various types. A graph can represent, for example, a social network that connects users to each other, a transport network like the metro where stations are connected to each other, or a brain with the billions of interacting neurons it contains. In recent years, the dynamic nature of these structures has been highlighted, as well as the importance of taking into account the temporal evolution of these networks to understand their functioning. While many concepts and algorithms have been developed on graphs to describe static network structures, much remains to be done to formalize and develop relevant algorithms to describe the dynamics of real networks. This thesis aims to better understand how massive graphs are structured in the real world, and to develop tools to extend our understanding to structures that evolve over time. It has been shown that these graphs have particular properties, which distinguish them from theoretical or randomly drawn graphs. Exploiting these properties then enables the design of algorithms to solve certain difficult problems much more quickly on these instances than in the general case. My PhD thesis focuses on cliques, which are groups of elements that are all connected to each other. We study the enumeration of cliques in static and temporal graphs and the detection of communities they enable. The communities of a graph are sets of vertices such that, within a community, the vertices interact strongly with each other, and little with the rest of the graph. Their study helps to understand the structural and functional properties of networks. We are evaluating our algorithms on massive real-world graphs, opening up new perspectives for understanding interactions within these networks. We first work on graphs, without taking into account the temporal component of interactions. We begin by using the clique percolation method of community detection, highlighting its limitations in memory, which prevent it from being applied to graphs that are too massive. By introducing an approximate problem-solving algorithm, we overcome this limitation. Next, we improve the enumeration of maximal cliques in the case of bipartite graphs. These correspond to interactions between groups of vertices of different types, e.g. links between people and viewed content, participation in events, etc. Next, we consider interactions that take place over time, using the link stream formalism. We seek to extend the algorithms presented in the first part, to exploit their advantages in the study of temporal interactions. We provide a new algorithm for enumerating maximal cliques in link streams, which is much more efficient than the state-of-the-art on massive datasets. Finally, we focus on communities in link streams by clique percolation, developing an extension of the method used on graphs. The results show a significant improvement over the state of the art, and we analyze the communities obtained to provide relevant information on the organization of temporal interactions in link streams. My PhD work has provided new insights into the study of massive real-world networks. This shows the importance of exploring the potential of graphs in a real-world context, and could contribute to the emergence of innovative solutions for the complex challenges of our modern society
Wang, Tian. "Abnormal detection in video streams via one-class learning methods." Thesis, Troyes, 2014. http://www.theses.fr/2014TROY0018/document.
One of the major research areas in computer vision is visual surveillance. The scientific challenge in this area includes the implementation of automatic systems for obtaining detailed information about the behavior of individuals and groups. Particularly, detection of abnormal individual movements requires sophisticated image analysis. This thesis focuses on the problem of the abnormal events detection, including feature descriptor design characterizing the movement information and one-class kernel-based classification methods. In this thesis, three different image features have been proposed: (i) global optical flow features, (ii) histograms of optical flow orientations (HOFO) descriptor and (iii) covariance matrix (COV) descriptor. Based on these proposed descriptors, one-class support vector machines (SVM) are proposed in order to detect abnormal events. Two online strategies of one-class SVM are proposed: The first strategy is based on support vector description (online SVDD) and the second strategy is based on online least squares one-class support vector machines (online LS-OC-SVM)
Merino, Pierrick. "Reproduction expérimentale du contact roue-rail à échelle réduite : Voies de formation des sources de défauts." Thesis, Lyon, 2019. http://www.theses.fr/2019LYSEI101.
The safety issue is still the main concern of railway network due to the development of railway transportation and the increase of the amount of passengers. The understanding of the origin of the rolling contact fatigue (RCF) defect, is one key to safety requirements. The White Etching Layer associated to the initiation of the squat defect is hardly recreated. The use of laboratory test bench enable the replication of the wheel-rail contact. Nevertheless, only a fraction of the characteristic parameters is taken into account and compromises are necessary. The test bench “Triboring” built at LaMCoS, fulfills a gap in the existing apparatus. The “roller on circular rail” design was chosen to fit the tribological behavior of wheel-rail contact, and replicate RCF defects. The production of a test bench required to relate and differentiate the measured data to the phenomena corresponding to the operating from the phenomena corresponding to wheel-rail contact. The bench was characterized with dynamic and cinematic analysis. The design of the sample was improved. The tribological behavior of the bench was optimized with the preparation of the initial surface of the samples and the production of a tribological “fuse”. This layer delays the speed accommodation by wear and benefit the shear of the first bodies and the formation a Tribological Transformation of Surface (TTS), as the White Etching Layer. The two different fuse layer created (Run-in and oxidized), induced a significant wear reduction. The tribological and metallurgical analysis of the surfaces and cuts of the sample, enabled to the explanation of the evolution of the microstructure of the rail close to the surface, submitted to various mechanical solicitations. The transformation of this microstructure led to the formation of white etching layer mechanically formed
Wilmet, Audrey. "Détection d'anomalies dans les flots de liens : combiner les caractéristiques structurelles et temporelles." Electronic Thesis or Diss., Sorbonne université, 2019. http://www.theses.fr/2019SORUS402.
A link stream is a set of links {(t, u, v)} in which a triplet (t, u, v) models the interaction between two entities u and v at time t. In many situations, data result from the measurement of interactions between several million of entities over time and can thus be studied through the link stream's formalism. This is the case, for instance, of phone calls, email exchanges, money transfers, contacts between individuals, IP traffic, online shopping, and many more. The goal of this thesis is the detection of sets of abnormal links in a link stream. In a first part, we design a method that constructs different contexts, a context being a set of characteristics describing the circumstances of an anomaly. These contexts allow us to find unexpected behaviors that are relevant, according to several dimensions and perspectives. In a second part, we design a method to detect anomalies in heterogeneous distributions whose behavior is constant over time, by comparing a sequence of similar heterogeneous distributions. We apply our methodological tools to temporal interactions coming from retweets of Twitter and IP traffic of MAWI group
Laby, Romain. "Détection et localisation d'anomalies dans des données hétérogènes en utilisant des modèles graphiques non orientés mixtes." Electronic Thesis or Diss., Paris, ENST, 2017. http://www.theses.fr/2017ENST0026.
This thesis revolves around an industrial need of Thales Système Aéroportés and the RBE2 combat radar equipping Dassault Rafale fighter aircraft. It develops a methodology for locating anomalies in heterogeneous data stream using a mixed, non-orientation and peer-to-peer graphical model. The data are a mixture of categorical and quantitative variables, and the model is learned from a data set that is assumed not to contain abnormal data. Anomaly localization algorithms use an adapted version of the CUSUM algorithm, whose decision function is based on the calculation of conditional likelihood ratios. This function allows the detection of variable anomalies per variable and the precise localization of the variables involved in the anomaly
Foulon, Lucas. "Détection d'anomalies dans les flux de données par structure d'indexation et approximation : Application à l'analyse en continu des flux de messages du système d'information de la SNCF." Thesis, Lyon, 2020. http://www.theses.fr/2020LYSEI082.
In this thesis, we propose methods to approximate an anomaly score in order to detect abnormal parts in data streams. Two main problems are considered in this context. Firstly, the handling of the high dimensionality of the objects describing the time series extracted from the raw streams, and secondly, the low computation cost required to perform the analysis on-the-fly. To tackle the curse of dimensionality, we have selected the CFOF anomaly score, that has been proposed recently and proven to be robust to the increase of the dimensionality. Our main contribution is then the proposition of two methods to quickly approximate the CFOF score of new objects in a stream. The first one is based on safe pruning and approximation during the exploration of object neighbourhood. The second one is an approximation obtained by the aggregation of scores computed in several subspaces. Both contributions complete each other and can be combined. We show on a reference benchmark that our proposals result in important reduction of the execution times, while providing approximations that preserve the quality of anomaly detection. Then, we present our application of these approaches within the SNCF information system. In this context, we have extended the existing monitoring modules by a new tool to help to detect abnormal behaviours in the real stream of messages within the SNCF communication system
Kammoun, Abderrahmen. "Enhancing Stream Processing and Complex Event Processing Systems." Thesis, Lyon, 2019. http://www.theses.fr/2019LYSES012.
As more and more connected objects and sensory devices are becoming part of our daily lives, the sea of high-velocity information flow is growing. This massive amount of data produced at high rates requires rapid insight to be useful in various applications such as the Internet of Things, health care, energy management, etc. Traditional data storage and processing techniques are proven inefficient. This gives rise to Data Stream Management and Complex Event Processing (CEP) systems.This thesis aims to provide optimal solutions for complex and proactive queries. Our proposed techniques, in addition to CPU and memory efficiency, enhance the capabilities of existing CEP systems by adding predictive feature through real-time learning. The main contributions of this thesis are as follows:We proposed various techniques to reduce the CPU and memory requirements of expensive queries. These operators result in exponential complexity both in terms of CPU and memory. Our proposed recomputation and heuristic-based algorithm reduce the costs of these operators. These optimizations are based on enabling efficient multidimensional indexing using space-filling curves and by clustering events into batches to reduce the cost of pair-wise joins.We designed a novel predictive CEP system that employs historical information to predict future complex events. We proposed a compressed index structure, range query processing techniques and an approximate summarizing technique over the historical space.The applicability of our techniques over the real-world problems presented has produced further customize-able solutions that demonstrate the viability of our proposed methods
Mazoyer, Béatrice. "Social Media Stories. Event detection in heterogeneous streams of documents applied to the study of information spreading across social and news media." Thesis, université Paris-Saclay, 2020. http://www.theses.fr/2020UPASC009.
Social Media, and Twitter in particular, has become a privileged source of information for journalists in recent years. Most of them monitor Twitter, in the search for newsworthy stories. This thesis aims to investigate and to quantify the effect of this technological change on editorial decisions. Does the popularity of a story affects the way it is covered by traditional news media, regardless of its intrinsic interest?To highlight this relationship, we take a multidisciplinary approach at the crossroads of computer science and economics: first, we design a novel approach to collect a representative sample of 70% of all French tweets emitted during an entire year. Second, we study different types of algorithms to automatically discover tweets that relate to the same stories. We test several vector representations of tweets, looking at both text and text-image representations, Third, we design a new method to group together Twitter events and media events. Finally, we design an econometric instrument to identify a causal effect of the popularity of an event on Twitter on its coverage by traditional media. We show that the popularity of a story on Twitter does have an effect on the number of articles devoted to it by traditional media, with an increase of about 1 article per 1000 additional tweets
Gaumont, Noé. "Groupes et Communautés dans les flots de liens : des données aux algorithmes." Thesis, Paris 6, 2016. http://www.theses.fr/2016PA066271/document.
Interactions are everywhere: in the contexts of face-to-face contacts, emails, phone calls, IP traffic, etc. In all of them, an interaction is characterized by two entities and a time interval: for instance, two individuals meet from 1pm to 3pm. We model them as link stream which is a set of quadruplets (b,e,u,v) where each quadruplet means that a link exists between u and v from time b to time e. In graphs, a community is a subset which is more densely connected than a reference. Within the link stream formalism, the notion of density and reference have to be redefined. Therefore, we study how to extend the notion of density for link streams. To this end, we use a real data set where a community structure is known. Then, we develop a method that finds automatically substream which are considered relevant. These substream, defined as subsets of links, are discovered by a classical community detection algorithm applied on the link stream the transformed into a static graph. A substream is considered relevant, if it is denser than the substreams which are close temporally and structurally. Thus, we deepen the notion of neighbourhood and reference in link streams. We apply our method on several real world interaction networks and we find relevant substream which would not have been found by existing methods. Finally, we discuss the generation of link streams having a given community structure and also a proper way to evaluate such community structure
Putina, Andrian. "Unsupervised anomaly detection : methods and applications." Electronic Thesis or Diss., Institut polytechnique de Paris, 2022. http://www.theses.fr/2022IPPAT012.
An anomaly (also known as outlier) is an instance that significantly deviates from the rest of the input data and being defined by Hawkins as 'an observation, which deviates so much from other observations as to arouse suspicions that it was generated by a different mechanism'. Anomaly detection (also known as outlier or novelty detection) is thus the machine learning and data mining field with the purpose of identifying those instances whose features appear to be inconsistent with the remainder of the dataset. In many applications, correctly distinguishing the set of anomalous data points (outliers) from the set of normal ones (inliers) proves to be very important. A first application is data cleaning, i.e., identifying noisy and fallacious measurement in a dataset before further applying learning algorithms. However, with the explosive growth of data volume collectable from various sources, e.g., card transactions, internet connections, temperature measurements, etc. the use of anomaly detection becomes a crucial stand-alone task for continuous monitoring of the systems. In this context, anomaly detection can be used to detect ongoing intrusion attacks, faulty sensor networks or cancerous masses.The thesis proposes first a batch tree-based approach for unsupervised anomaly detection, called 'Random Histogram Forest (RHF)'. The algorithm solves the curse of dimensionality problem using the fourth central moment (aka kurtosis) in the model construction while boasting linear running time. A stream based anomaly detection engine, called 'ODS', that leverages DenStream, an unsupervised clustering technique is presented subsequently and finally Automated Anomaly Detection engine which alleviates the human effort required when dealing with several algorithm and hyper-parameters is presented as last contribution
Bouguelia, Mohamed-Rafik. "Classification et apprentissage actif à partir d'un flux de données évolutif en présence d'étiquetage incertain." Thesis, Université de Lorraine, 2015. http://www.theses.fr/2015LORR0034/document.
This thesis focuses on machine learning for data classification. To reduce the labelling cost, active learning allows to query the class label of only some important instances from a human labeller.We propose a new uncertainty measure that characterizes the importance of data and improves the performance of active learning compared to the existing uncertainty measures. This measure determines the smallest instance weight to associate with new data, so that the classifier changes its prediction concerning this data. We then consider a setting where the data arrives continuously from an infinite length stream. We propose an adaptive uncertainty threshold that is suitable for active learning in the streaming setting and achieves a compromise between the number of classification errors and the number of required labels. The existing stream-based active learning methods are initialized with some labelled instances that cover all possible classes. However, in many applications, the evolving nature of the stream implies that new classes can appear at any time. We propose an effective method of active detection of novel classes in a multi-class data stream. This method incrementally maintains a feature space area which is covered by the known classes, and detects those instances that are self-similar and external to that area as novel classes. Finally, it is often difficult to get a completely reliable labelling because the human labeller is subject to labelling errors that reduce the performance of the learned classifier. This problem was solved by introducing a measure that reflects the degree of disagreement between the manually given class and the predicted class, and a new informativeness measure that expresses the necessity for a mislabelled instance to be re-labeled by an alternative labeller
Vervaet, Arthur. "Automated Log-Based Anomaly Detection within Cloud Computing Infrastructures." Electronic Thesis or Diss., Sorbonne université, 2023. http://www.theses.fr/2023SORUS548.
Cloud computing aims to optimize resource utilization while accommodating a large user base and elastic services. Within this context, cloud computing platforms bear the responsibility of managing their customers’ infrastructure. The management of an everexpanding number of IT resources poses a significant challenge. In this study, conducted in collaboration with 3DS OUTSCALE, a French public cloud provider, we investigate the potential of log data as a valuable source for automated anomaly detection within cloud computing platforms. Logs serve as a widely utilized information source for various purposes, including monitoring, diagnosing, performance evaluation, and maintenance. These logs are generated during runtime and provide insights into the current state of a system. However, achieving automated real-time anomaly detection based on log data remains a complex undertaking. The intricate nature of cloud computing platforms must be duly considered. Extracting relevant information from a multitude of logging sources and accounting for frequent code base evolution poses challenges and introduces the potential for errors. Furthermore, establishing log relationships within such systems is often an insurmountable task. Log parsing solutions aim to extract variables from the template of log messages. Our first contribution involves a comprehensive study of two state-of-the-art log parsing methods, investigating the impact of hyperparameter tuning and preprocessing on their accuracy. Given the laborious nature of labeling logs related to a cloud computing platform, we sought to identify potential generic values that enable accurate parsing across diverse scenarios. However, our research reveals the infeasibility of finding such requirements, thereby emphasizing the necessity for more robust parsing approaches. Our second contribution introduces USTEP, an innovative online log parsing approach that surpasses existing methods in terms of accuracy, efficiency, and robustness. Notably, USTEP achieves a constant worst-case parsing time complexity, distinguishing it from its predecessors for which the number of already detected templates is to be taken into account. Through a comparative analysis of five online log parsers using 13 open-source datasets and one derived from 3DS OUTSCALE systems, we demonstrate the superior performance of USTEP. Furthermore, we propose USTEP-UP, an architecture that enables the distributed execution of multiple USTEP instances. Our third contribution presents Monilog, a system architecture designed for automated log-based anomaly detection within log data streams. Monilog leverages model/metric pairs to predict log traffic patterns within a system and detect anomalies by identifying deviations in system behavior. Monilog forecasting models are powered by the recent advances in the deep learning field and is able to generate comprehensive reports that highlight the relevant system components and the associated applications. We implemented an instance of Monilog at cloud scale and conducted experimental analyses to evaluate its ability to forecast anomalous events, such as servers crashes resulting from virtualization issues. The results obtained strongly support our hypothesis regarding the utility of logs in detecting and predicting abnormal events. Our Monilog implementation successfully identified abnormal periods and provided valuable insights into the applications involved. With Monilog, we demonstrate the value of logs in predicting anomalies in such environments and provide a flexible architecture for future study. Our work on the parsing field with the proposal of USTEP and USTEP-UP not only provides us with additional information for building anomaly detection models but also has potential benefits for other log mining applications
Gaumont, Noé. "Groupes et Communautés dans les flots de liens : des données aux algorithmes." Electronic Thesis or Diss., Paris 6, 2016. http://www.theses.fr/2016PA066271.
Interactions are everywhere: in the contexts of face-to-face contacts, emails, phone calls, IP traffic, etc. In all of them, an interaction is characterized by two entities and a time interval: for instance, two individuals meet from 1pm to 3pm. We model them as link stream which is a set of quadruplets (b,e,u,v) where each quadruplet means that a link exists between u and v from time b to time e. In graphs, a community is a subset which is more densely connected than a reference. Within the link stream formalism, the notion of density and reference have to be redefined. Therefore, we study how to extend the notion of density for link streams. To this end, we use a real data set where a community structure is known. Then, we develop a method that finds automatically substream which are considered relevant. These substream, defined as subsets of links, are discovered by a classical community detection algorithm applied on the link stream the transformed into a static graph. A substream is considered relevant, if it is denser than the substreams which are close temporally and structurally. Thus, we deepen the notion of neighbourhood and reference in link streams. We apply our method on several real world interaction networks and we find relevant substream which would not have been found by existing methods. Finally, we discuss the generation of link streams having a given community structure and also a proper way to evaluate such community structure
Belghaouti, Fethi. "Interopérabilité des systèmes distribués produisant des flux de données sémantiques au profit de l'aide à la prise de décision." Thesis, Université Paris-Saclay (ComUE), 2017. http://www.theses.fr/2017SACLL003.
Internet is an infinite source of data coming from sources such as social networks or sensors (home automation, smart city, autonomous vehicle, etc.). These heterogeneous and increasingly large data can be managed through semantic web technologies, which propose to homogenize, link these data and reason above them, and data flow management systems, which mainly address the problems related to volume, volatility and continuous querying. The alliance of these two disciplines has seen the growth of semantic data stream management systems also called RSP (RDF Stream Processing Systems). The objective of this thesis is to allow these systems, via new approaches and "low cost" algorithms, to remain operational, even more efficient, even for large input data volumes and/or with limited system resources.To reach this goal, our thesis is mainly focused on the issue of "Processing semantic data streamsin a context of computer systems with limited resources". It directly contributes to answer the following research questions : (i) How to represent semantic data stream ? And (ii) How to deal with input semantic data when their rates and/or volumes exceed the capabilities of the target system ?As first contribution, we propose an analysis of the data in the semantic data streams in order to consider a succession of star graphs instead of just a success of andependent triples, thus preserving the links between the triples. By using this approach, we significantly impoved the quality of responses of some well known sampling algoithms for load-shedding. The analysis of the continuous query allows the optimisation of this solution by selection the irrelevant data to be load-shedded first. In the second contribution, we propose an algorithm for detecting frequent RDF graph patterns in semantic data streams.We called it FreGraPaD for Frequent RDF Graph Patterns Detection. It is a one pass algorithm, memory oriented and "low-cost". It uses two main data structures : A bit-vector to build and identify the RDF graph pattern, providing thus memory space optimization ; and a hash-table for storing the patterns.The third contribution of our thesis consists of a deterministic load-shedding solution for RSP systems, called POL (Pattern Oriented Load-shedding for RDF Stream Processing systems). It uses very low-cost boolean operators, that we apply on the built binary patterns of the data and the continuous query inorder to determine which data is not relevant to be ejected upstream of the system. It guarantees a recall of 100%, reduces the system load and improves response time. Finally, in the fourth contribution, we propose Patorc (Pattern Oriented Compression for RSP systems). Patorc is an online compression toolfor RDF streams. It is based on the frequent patterns present in RDF data streams that factorizes. It is a data lossless compression solution whith very possible querying without any need to decompression.This thesis provides solutions that allow the extension of existing RSP systems and makes them able to scale in a bigdata context. Thus, these solutions allow the RSP systems to deal with one or more semantic data streams arriving at different speeds, without loosing their response quality while ensuring their availability, even beyond their physical limitations. The conducted experiments, supported by the obtained results show that the extension of existing systems with the new solutions improves their performance. They illustrate the considerable decrease in their engine’s response time, increasing their processing rate threshold while optimizing the use of their system resources
Bouguelia, Mohamed-Rafik. "Classification et apprentissage actif à partir d'un flux de données évolutif en présence d'étiquetage incertain." Electronic Thesis or Diss., Université de Lorraine, 2015. http://www.theses.fr/2015LORR0034.
This thesis focuses on machine learning for data classification. To reduce the labelling cost, active learning allows to query the class label of only some important instances from a human labeller.We propose a new uncertainty measure that characterizes the importance of data and improves the performance of active learning compared to the existing uncertainty measures. This measure determines the smallest instance weight to associate with new data, so that the classifier changes its prediction concerning this data. We then consider a setting where the data arrives continuously from an infinite length stream. We propose an adaptive uncertainty threshold that is suitable for active learning in the streaming setting and achieves a compromise between the number of classification errors and the number of required labels. The existing stream-based active learning methods are initialized with some labelled instances that cover all possible classes. However, in many applications, the evolving nature of the stream implies that new classes can appear at any time. We propose an effective method of active detection of novel classes in a multi-class data stream. This method incrementally maintains a feature space area which is covered by the known classes, and detects those instances that are self-similar and external to that area as novel classes. Finally, it is often difficult to get a completely reliable labelling because the human labeller is subject to labelling errors that reduce the performance of the learned classifier. This problem was solved by introducing a measure that reflects the degree of disagreement between the manually given class and the predicted class, and a new informativeness measure that expresses the necessity for a mislabelled instance to be re-labeled by an alternative labeller
Belghaouti, Fethi. "Interopérabilité des systèmes distribués produisant des flux de données sémantiques au profit de l'aide à la prise de décision." Electronic Thesis or Diss., Université Paris-Saclay (ComUE), 2017. http://www.theses.fr/2017SACLL003.
Internet is an infinite source of data coming from sources such as social networks or sensors (home automation, smart city, autonomous vehicle, etc.). These heterogeneous and increasingly large data can be managed through semantic web technologies, which propose to homogenize, link these data and reason above them, and data flow management systems, which mainly address the problems related to volume, volatility and continuous querying. The alliance of these two disciplines has seen the growth of semantic data stream management systems also called RSP (RDF Stream Processing Systems). The objective of this thesis is to allow these systems, via new approaches and "low cost" algorithms, to remain operational, even more efficient, even for large input data volumes and/or with limited system resources.To reach this goal, our thesis is mainly focused on the issue of "Processing semantic data streamsin a context of computer systems with limited resources". It directly contributes to answer the following research questions : (i) How to represent semantic data stream ? And (ii) How to deal with input semantic data when their rates and/or volumes exceed the capabilities of the target system ?As first contribution, we propose an analysis of the data in the semantic data streams in order to consider a succession of star graphs instead of just a success of andependent triples, thus preserving the links between the triples. By using this approach, we significantly impoved the quality of responses of some well known sampling algoithms for load-shedding. The analysis of the continuous query allows the optimisation of this solution by selection the irrelevant data to be load-shedded first. In the second contribution, we propose an algorithm for detecting frequent RDF graph patterns in semantic data streams.We called it FreGraPaD for Frequent RDF Graph Patterns Detection. It is a one pass algorithm, memory oriented and "low-cost". It uses two main data structures : A bit-vector to build and identify the RDF graph pattern, providing thus memory space optimization ; and a hash-table for storing the patterns.The third contribution of our thesis consists of a deterministic load-shedding solution for RSP systems, called POL (Pattern Oriented Load-shedding for RDF Stream Processing systems). It uses very low-cost boolean operators, that we apply on the built binary patterns of the data and the continuous query inorder to determine which data is not relevant to be ejected upstream of the system. It guarantees a recall of 100%, reduces the system load and improves response time. Finally, in the fourth contribution, we propose Patorc (Pattern Oriented Compression for RSP systems). Patorc is an online compression toolfor RDF streams. It is based on the frequent patterns present in RDF data streams that factorizes. It is a data lossless compression solution whith very possible querying without any need to decompression.This thesis provides solutions that allow the extension of existing RSP systems and makes them able to scale in a bigdata context. Thus, these solutions allow the RSP systems to deal with one or more semantic data streams arriving at different speeds, without loosing their response quality while ensuring their availability, even beyond their physical limitations. The conducted experiments, supported by the obtained results show that the extension of existing systems with the new solutions improves their performance. They illustrate the considerable decrease in their engine’s response time, increasing their processing rate threshold while optimizing the use of their system resources
"Détection du stress hydrique par thermographie infrarouge. Application à la culture de la pomme de terre." Thesis, Université Laval, 2004. http://www.theses.ulaval.ca/2004/22198/22198.pdf.
Badouard, Carine. "Les lésions des acides nucléiques: détection par CLHP-SM/SM dans les milieux biologiques humains et intérêt comme biomarqueurs du stress oxydant et de l'inflammation." Phd thesis, 2006. http://tel.archives-ouvertes.fr/tel-00134563.