Dissertations / Theses on the topic 'Prédiction automatique'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Prédiction automatique.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
Elloumi, Zied. "Prédiction de performances des systèmes de Reconnaissance Automatique de la Parole." Thesis, Université Grenoble Alpes (ComUE), 2019. http://www.theses.fr/2019GREAM005/document.
Full textIn this thesis, we focus on performance prediction of automatic speech recognition (ASR) systems.This is a very useful task to measure the reliability of transcription hypotheses for a new data collection, when the reference transcription is unavailable and the ASR system used is unknown (black box).Our contribution focuses on several areas: first, we propose a heterogeneous French corpus to learn and evaluate ASR prediction systems.We then compare two prediction approaches: a state-of-the-art (SOTA) performance prediction based on engineered features and a new strategy based on learnt features using convolutional neural networks (CNNs).While the joint use of textual and signal features did not work for the SOTA system, the combination of inputs for CNNs leads to the best WER prediction performance. We also show that our CNN prediction remarkably predicts the shape of the WER distribution on a collection of speech recordings.Then, we analyze factors impacting both prediction approaches. We also assess the impact of the training size of prediction systems as well as the robustness of systems learned with the outputs of a particular ASR system and used to predict performance on a new data collection.Our experimental results show that both prediction approaches are robust and that the prediction task is more difficult on short speech turns as well as spontaneous speech style.Finally, we try to understand which information is captured by our neural model and its relation with different factors.Our experiences show that intermediate representations in the network automatically encode information on the speech style, the speaker's accent as well as the broadcast program type.To take advantage of this analysis, we propose a multi-task system that is slightly more effective on the performance prediction task
Kawala, François. "Prédiction de l'activité dans les réseaux sociaux." Thesis, Université Grenoble Alpes (ComUE), 2015. http://www.theses.fr/2015GREAM021/document.
Full textThis dissertation is devoted to a social-media-mining problem named the activity-prediction problem. In this problem one aims to predict the number of user-generated-contents that will be created about a topic in the near future. The user-generated-contents that belong to a topic are not necessary related to each other.In order to study the activity-prediction problem without referring directly to a particular social-media, a generic framework is proposed. This generic framework allows to describe various social-media in a unified way. With this generic framework the activityprediction problem is defined independently of an actual social-media. Three examples are provided to illustrate how this generic framework describes social-media. Three defi- nitions of the activity-prediction problem are proposed. Firstly the magnitude prediction problem defines the activity-prediction as a regression problem. With this definition one aims to predict the exact activity of a topic. Secondly, the buzz classification problem defines the activity-prediction as a binary classification problem. With this definition one aims to predict if a topic will have an activity burst of a predefined amplitude. Thirdly the rank prediction problem defines the activity-prediction as a learning-to-rank problem. With this definition one aims to rank the topics accordingly to theirs future activity-levels. These three definitions of the activity prediction problem are tackled with state-of-the-art machine learning approaches applied to generic features. Indeed, these features are defined with the help of the generic framework. Therefore these features are easily adaptable to various social-media. There are two types of features. Firstly the features which describe a single topic. Secondly the features which describe the interplay between two topics.Our ability to predict the activity is tested against an industrial-size multilingual dataset. The data has been collected during 51 weeks. Two sources of data were used: Twitter and a bulletin-board-system. The collected data contains three languages: English, French and German. More than five hundred millions user-generated-contents were captured. Most of these user-generated-contents are related to computer hardware, video games, and mobile telephony. The data collection necessitated the implementation of a daily routine. The data was prepared so that commercial-contents and technical failure are not sources of noise. A cross-validation method that takes into account the time of observations is used. In addition an unsupervised method to extract buzz candidates is proposed. Indeed the training-sets are very ill-balanced for the buzz classification problem, and it is necessary to preselect buzz candidates. The activity-prediction problems are studied within two different experimental settings. The first experimental setting includes data from Twitter and the bulletin-board-system, on a long time-scale, and with three different languages. The second experimental setting is dedicated specifically to Twitter. This second experiment aims to increase the reproducibility of experiments as much as possible. Hence, this experimental setting includes user-generated-contents collected with respect to a list of unambiguous English terms. In addition the observation are restricted to ten consecutive weeks. Hence the risk of unannounced change in the public API of Twitter is minimized
Hmamouche, Youssef. "Prédiction des séries temporelles larges." Electronic Thesis or Diss., Aix-Marseille, 2018. http://www.theses.fr/2018AIXM0480.
Full textNowadays, storage and data processing systems are supposed to store and process large time series. As the number of variables observed increases very rapidly, their prediction becomes more and more complicated, and the use of all the variables poses problems for classical prediction models.Univariate prediction models are among the first models of prediction. To improve these models, the use of multiple variables has become common. Thus, multivariate models and become more and more used because they consider more information.With the increase of data related to each other, the application of multivariate models is also questionable. Because the use of all existing information does not necessarily lead to the best predictions. Therefore, the challenge in this situation is to find the most relevant factors among all available data relative to a target variable.In this thesis, we study this problem by presenting a detailed analysis of the proposed approaches in the literature. We address the problem of prediction and size reduction of massive data. We also discuss these approaches in the context of Big Data.The proposed approaches show promising and very competitive results compared to well-known algorithms, and lead to an improvement in the accuracy of the predictions on the data used.Then, we present our contributions, and propose a complete methodology for the prediction of wide time series. We also extend this methodology to big data via distributed computing and parallelism with an implementation of the prediction process proposed in the Hadoop / Spark environment
Popov, Mihail. "Décomposition automatique des programmes parallèles pour l'optimisation et la prédiction de performance." Thesis, Université Paris-Saclay (ComUE), 2016. http://www.theses.fr/2016SACLV087/document.
Full textIn high performance computing, benchmarks evaluate architectures, compilers and optimizations. Standard benchmarks are mostly issued from the industrial world and may have a very long execution time. So, evaluating a new architecture or an optimization is costly. Most of the benchmarks are composed of independent kernels. Usually, users are only interested by a small subset of these kernels. To get faster and easier local optimizations, we should find ways to extract kernels as standalone executables. Also, benchmarks have redundant computational kernels. Some calculations do not bring new informations about the system that we want to study, despite that we measure them many times. By detecting similar operations and removing redundant kernels, we can reduce the benchmarking cost without loosing information about the system. This thesis proposes a method to automatically decompose applications into small kernels called codelets. Each codelet is a standalone executable that can be replayed in different execution contexts to evaluate them. This thesis quantifies how much the decomposition method accelerates optimization and benchmarking processes. It also quantify the benefits of local optimizations over global optimizations. There are many related works which aim to enhance the benchmarking process. In particular, we note machine learning approaches and sampling techniques. Decomposing applications into independent pieces is not a new idea. It has been successfully applied on sequential codes. In this thesis we extend it to parallel programs. Evaluating scalability or new micro-architectures is 25× faster with codelets than with full application executions. Codelets predict the execution time with an accuracy of 94% and find local optimizations that outperform the best global optimization up to 1.06×
Colombet-Madinier, Isabelle. "Aspects méthodologiques de la prédiction du risque cardiovasculaire : apports de l'apprentissage automatique." Paris 6, 2002. http://www.theses.fr/2002PA066083.
Full textHmamouche, Youssef. "Prédiction des séries temporelles larges." Thesis, Aix-Marseille, 2018. http://www.theses.fr/2018AIXM0480.
Full textNowadays, storage and data processing systems are supposed to store and process large time series. As the number of variables observed increases very rapidly, their prediction becomes more and more complicated, and the use of all the variables poses problems for classical prediction models.Univariate prediction models are among the first models of prediction. To improve these models, the use of multiple variables has become common. Thus, multivariate models and become more and more used because they consider more information.With the increase of data related to each other, the application of multivariate models is also questionable. Because the use of all existing information does not necessarily lead to the best predictions. Therefore, the challenge in this situation is to find the most relevant factors among all available data relative to a target variable.In this thesis, we study this problem by presenting a detailed analysis of the proposed approaches in the literature. We address the problem of prediction and size reduction of massive data. We also discuss these approaches in the context of Big Data.The proposed approaches show promising and very competitive results compared to well-known algorithms, and lead to an improvement in the accuracy of the predictions on the data used.Then, we present our contributions, and propose a complete methodology for the prediction of wide time series. We also extend this methodology to big data via distributed computing and parallelism with an implementation of the prediction process proposed in the Hadoop / Spark environment
Herry, Sébastien. "Détection automatique de langue par discrimination d'experts." Paris 6, 2007. http://www.theses.fr/2007PA066101.
Full textThe purpose of the presented work in this memoir is to automatically detect language in audio stream. For this we suggest a model which, like bilingual expert, done an discrimination by language pair with only acoustic information. The system have constraint : Operating in real time, Use database without phonetic information, Able to add a new language without retrain all the model In a first time we have done an Automatic language detection system derived from the stat of the art. The results obtained by this system are used as reference for the rest of memoir, and we compare those results with the results obtained by the developed model. In a first time, we propose a set of discriminator, by pair of language, based on neural network. The treatment is done on the whole duration of speech segment. The results of these discriminators are fused to create de detection. This model has a patent. We have study more precisely the influence of different parameter as the number of locator, the variation intra and inter corpus or the hardiness. Next we have compared the proposed modelling based on discrimination, with modelling auto regressive or predictive. This system has been tested with our participation of the international campaign organised by NIST in December 2005. To conclude on this campaign where 17 international teams have participated, we have proposed several improvements as: A normalisation of database, A modification of speaker database for learning only, Increase scores with segment duration. To conclude, the system proposed fulfils the constraints because the system is real time, and use only acoustic information. More over the system is more efficient than the derived model from the stat of the art. At last the model is hardiness for noise, for unknown language, for new evaluation database
Kashnikov, Yuriy. "Une approche holistique pour la prédiction des optimisations du compilateur par apprentissage automatique." Versailles-St Quentin en Yvelines, 2013. http://www.theses.fr/2013VERS0047.
Full textEffective compiler optimizations can greatly improve applications performance. These optimizations are numerous and can be applied in any order. Compilers select these optimizations using solutions driven by heuristics which may degrade programs performance. Therefore, developers resort to the tedious manual search for the best optimizations. Combinatorial search space makes this effort intractable and one can easily fall into a local minimum and miss the best combination. This thesis develops a holistic approach to improve applications performance with compiler optimizations and machine learning. A combination of static loop analysis and statistical learning is used to analyze a large corpus of loops and reveal good potential for compiler optimizations. Milepost GCC, a machine-learning based compiler, is applied to optimize benchmarks and an industrial database application. It uses function level static features and classification algorithms to predict a good sequence of optimizations. While Milepost GCC can mispredict the best optimizations, in general it obtains considerable speedups and outperforms state-of-the-art compiler heuristics. The culmination of this thesis is the ULM meta-optimization framework. ULM characterizes applications at different levels with static code features and hardware performance counters and finds the most important combination of program features. By selecting among three classification algorithms and tuning their parameters, ULM builds a sophisticated predictor that can outperform existing solutions. As a result, the ULM framework predicted correctly the best sequence of optimizations sequence in 92% of cases
Hue, Martial. "Méthodes à noyau pour l'annotation automatique et la prédiction d'interaction de structures de protéine." Paris 7, 2011. http://www.theses.fr/2011PA077151.
Full textAs large quantities of protein 3D structures are now routinely solved, there is a need for computational tools to automatically annotate protein structures. In this thesis, we investigate several machine learning approaches for this purpose, based on the popular support vector machine (SVM) algorithm. Indeed, the SVM offers several possibilities to overcome the complexity of protein structures, and their interactions. We propose to solve both issues by investigating new positive definite kernels. First, a kernel function for the annotation of protein structures is devised. The kernel is based on a similarity measure called MAMMOTH. Classification tasks corresponding to Enzyme Classification (EC), Structural Classification of Proteins (SCOP), and Gene Ontology (GO) annotation, show that the MAMMOTH kernel significantly outperforms other choices of kernels for protein structures and classifiers. Second, we design a kernel in the context of binary supervised prediction of objects with a specific structure, namely pairs of general objects. The problem of the inference of missing edges in a protein-protein interaction network may be cast in this context. Our results on three benchmarks of interaction between protein structures suggest that the Metric Learning Pairwise Kernel (MLPK), in combination with the MAMMOTH kernel, yield the best performance. Lastly, we introduce a new and efficient learning method for the supervised prediction of protein interaction. A pairwise kernel method is motivated by two previous methods, the Tensor Product Pairwise Kernel (TPPK) and the local model. The connection between the approaches is explicited and the two methods are formulated in a new common framework, that yields to natural generalization by an interpolation
Grenet, Ingrid. "De l’utilisation des données publiques pour la prédiction de la toxicité des produits chimiques." Thesis, Université Côte d'Azur (ComUE), 2019. http://www.theses.fr/2019AZUR4050.
Full textCurrently, chemical safety assessment mostly relies on results obtained in in vivo studies performed in laboratory animals. However, these studies are costly in term of time, money and animals used and therefore not adapted for the evaluation of thousands of compounds. In order to rapidly screen compounds for their potential toxicity and prioritize them for further testing, alternative solutions are envisioned such as in vitro assays and computational predictive models. The objective of this thesis is to evaluate how the public data from ToxCast and ToxRefDB can allow the construction of this type of models in order to predict in vivo effects induced by compounds, only based on their chemical structure. To do so, after data pre-processing, we first focus on the prediction of in vitro bioactivity from chemical structure and then on the prediction of in vivo effects from in vitro bioactivity data. For the in vitro bioactivity prediction, we build and test various models based on compounds’ chemical structure descriptors. Since learning data are highly imbalanced in favor of non-toxic compounds, we test a data augmentation technique and show that it improves models’ performances. We also perform a largescale study to predict hundreds of in vitro assays from ToxCast and show that the stacked generalization ensemble method leads to reliable models when used on their applicability domain. For the in vivo effects prediction, we evaluate the link between results from in vitro assays targeting pathways known to induce endocrine effects and in vivo effects observed in endocrine organs during longterm studies. We highlight that, unexpectedly, these assays are not predictive of the in vivo effects, which raises the crucial question of the relevance of in vitro assays. We thus hypothesize that the selection of assays able to predict in vivo effects should be based on complementary information such as, in particular, mechanistic data
Grivolla, Jens. "Apprentissage et décision automatique en recherche documentaire : prédiction de difficulté de requêtes et sélection de modèle de recherche." Avignon, 2006. http://www.theses.fr/2006AVIG0142.
Full textThis thesis is centered around the subject of information retrieval, with a focus on those queries that are particularly difficult to handle for current retrieval systems. In the application and evaluation settings we were concerned with, a user expresses his information need as a natural language query. There are different approaches for treating those queries, but current systems typically use a single approach for all queries, without taking into account the specific properties of each query. However, it has been shown that the performance of one strategy relative to another can vary greatly depending on the query. We have approached this problem by proposing methods that will permit to automatically identify those queries that will pose particular difficulties to the retrieval system, in order to allow for a specific treatment. This research topic was very new and barely starting to be explored at the beginning of my work, but has received much attention these last years. We have developed a certain number of quality predictor functions that obtain results comparable to those published recently by other research teams. However, the ability of individual predictors to accurately classify queries by their level of difficulty remains rather limited. The major particularity and originality of our work lies in the combination of those different measures. Using methods of automatic classification with corpus-based training, we have been able to obtain quite reliable predictions, on the basis of measures that individually are far less discriminant. We have also adapted our approach to other application settings, with very encouraging results. We have thus developed a method for the selective application of query expansion techniques, as well as the selection of the most appropriate retrieval model for each query
Salaün, Achille. "Prédiction d'alarmes dans les réseaux via la recherche de motifs spatio-temporels et l'apprentissage automatique." Electronic Thesis or Diss., Institut polytechnique de Paris, 2021. http://www.theses.fr/2021IPPAS010.
Full textNowadays, telecommunication networks occupy a central position in our world. Indeed, they allow to share worldwide a huge amount of information. Networks are however complex systems, both in size and technological diversity. Therefore, it makes their management and reparation more difficult. In order to limit the negative impact of such failures, some tools have to be developed to detect a failure whenever it occurs, analyse its root causes to solve it efficiently, or even predict this failure as prevention is better than cure. In this thesis, we mainly focus on these two last problems. To do so, we use files, called alarm logs, storing all the alarms that have been emitted by the system. However, these files are generally noisy and verbose: an operator managing a network needs tools able to extract and handle in an interpretable manner the causal relationships inside a log. In this thesis, we followed two directions. First, we have inspired from pattern matching techniques: similarly to the Ukkonen’s algorithm, we build online a structure, called DIG-DAG, that stores all the potential causal relationships between the events of a log. Moreover, we introduce a query system to exploit our DIG-DAG structure. Finally, we show how our solution can be used for root cause analysis. The second approach is a generative approach for the prediction of time series. In particular, we compare two well-known models for this task: recurrent neural nets on the one hand, hidden Markov models on the other hand. Here, we compare analytically the expressivity of these models by encompassing them into a probabilistic model, called GUM
Kallas, Maya. "Méthodes à noyaux en reconnaissance de formes, prédiction et classification : applications aux biosignaux." Troyes, 2012. http://www.theses.fr/2012TROY0026.
Full textThe proliferation of kernel methods lies essentially on the kernel trick, which induces an implicit nonlinear transformation with reduced computational cost. Still, the inverse transformation is often necessary. The resolution of this so-called pre-image problem enables new fields of applications of these methods. The main purpose of this thesis is to show that recent advances in statistical learning theory provide relevant solutions to several issues raised in signal and image processing. The first part focuses on the pre-image problem, and on solutions with constraints imposed by physiology. The non-negativity is probably the most commonly stated constraints when dealing with natural signals and images. Nonnegativity constraints on the result, as well as on the additivity of the contributions, are studied. The second part focuses on time series analysis according to a predictive approach. Autoregressive models are developed in the transformed space, while the prediction requires solving the pre-image problem. Two kernelbased predictive models are considered: the first one is derived by solving a least-squares problem, and the second one by providing the adequate Yule-Walker equations. The last part deals with the classification task for electrocardiograms, in order to detect anomalies. Detection and multi-class classification are explored in the light of support vector machines and self-organizing maps
Temanni, Mohamed-Ramzi. "Combinaison de sources de données pour l'amélioration de la prédiction en apprentissage : une application à la prédiction de la perte de poids chez l'obèse à partir de données transcriptomiques et cliniques." Paris 6, 2009. https://tel.archives-ouvertes.fr/tel-00814513.
Full textPham, Duc-Thinh. "Prédiction de trajectoire et avis de résolution de conflits de trafic aérien basée sur l’apprentissage automatique." Thesis, Paris Sciences et Lettres (ComUE), 2019. http://www.theses.fr/2019PSLEP027.
Full textThe increasing in traffic demand has strained air traffic control system and controllers which lead to the need of novel and efficient conflict detection and resolution advisory. In the scope of this thesis, we concentrate on studying challenges in conflict detection and resolution by using machine learning approaches. We have attempted to learn and predict controller behaviors from data using Random Forest. We also propose a novel approach for probabilistic conflict detection by using Heteroscedastic Gaussian Process as predictive models and Bayesian Optimization for probabilistic conflict detection algorithm. Finally, we propose an artificial intelligent agent that is capable of resolving conflicts, in the presence of traffic and uncertainty. The conflict resolution task is formulated as a decision-making problem in large and complex action space, which is applicable for employing reinforcement learning algorithm. Our work includes the development of a learning environment, scenario state representation, reward function, and learning algorithm. Machine learning methods have showed their advantages and potential in conflict detection and resolution related challenges. However, more studies would be conducted to improve their performances such as airspace network representation, multi-agent reinforcement learning or controller's strategy reconstruction from data
Ouliz, Rhita. "Prédiction de la variabilité spatiale de la disponibilité en biomasse résiduelle à l'aide de l'apprentissage automatique." Master's thesis, Université Laval, 2017. http://hdl.handle.net/20.500.11794/27789.
Full textResidual forest biomass is the woody biomass left over on the forest floor after harvesting. This biomass can be used as a source of renewable energy, at a price that may be, under certain conditions, competitive relative to other energy sources. The success of the use of residual forest biomass depends in part on an effective management of its supply chain. Thus, the risk management of supply disruption of residual forest biomass is essential to ensure the potential for expansion of a customer’s distribution network. This project aims to improve the supply chain profitability of residual forest biomass through effective management of sources of error related to the estimation of the availability of biomass. This is the estimation of the spatial variability of residual biomass with acceptable accuracy by using machine learning techniques. Machine learning is an attempt to replicate the concept of learning. It consists to design algorithms capable to learn from examples or samples in order to predict the values of targets In our case study, the KNN method will allow us to estimate residual biomass of the target area units (polygons) from the k nearest neighbour plots. To this effect, we will estimate initially the spatial variability in the availability of residual biomass using the machine learning method KNN (k nearest neighbours). We then determine the error of our estimation using a bootstrap method. Finally, we will develop the location of the residual forest biomass quantity taking into account the estimation error. The estimation results obtained in the framework of this research indicate an accuracy of 59,5 % to 71 % centred around 65,4 % with an estimation error of 29 % to 34,5 %. Our methodology has yielded relevant results compared with the study of Bernier et al. (2010) which has had accuracy of estimation equal to 19% of forest biomass volume using the KNN method. The use of this method may also be relevant for estimating the commercial forest biomass and for the prediction of forest biomass of each tree species.
Michalon, Olivier. "Modèles statistiques pour la prédiction de cadres sémantiques." Thesis, Aix-Marseille, 2017. http://www.theses.fr/2017AIXM0221/document.
Full textIn natural language processing, each analysis step has improved the way in which language can be modeled by machines. Another step of analysis still poorly mastered resides in semantic parsing. This type of analysis can provide information which would allow for many advances, such as better human-machine interactions or more reliable translations. There exist several types of meaning representation structures, such as PropBank, AMR and FrameNet. FrameNet corresponds to the frame semantic framework whose theory has been described by Charles Fillmore (1971). In this theory, each prototypical situation and each different elements involved are represented in such a way that two similar situations are represented by the same object, called a semantic frame. The work that we will describe here follows the work already developed for machine prediction of frame semantic representations. We will present four prediction systems, and each one of them allowed to validate another hypothesis on the necessary properties for effective prediction. We will show that semantic parsing can also be improved by providing prediction models with refined information as input of the system, with firstly a syntactic analysis where deep links are made explicit and secondly vectorial representations of the vocabulary learned beforehand
Raynaut, William. "Perspectives de méta-analyse pour un environnement d'aide à la simulation et prédiction." Thesis, Toulouse 3, 2018. http://www.theses.fr/2018TOU30005/document.
Full textThe emergence of the big data phenomenon has led to increasing demands in data analysis, which most often are conducted by other domains experts with little experience in data science. We then consider this important demand in intelligent assistance to data analysis, which receives an increasing attention from the scientific community. The first takes on the subject often possessing similar shortcomings, we propose to address it through new processes of meta-analysis. No evaluation standard having yet been set in this relatively new domain, we first propose a meta-analysis evaluation framework that will allow us to test and compare the developed methods. In order to open new approaches of meta-analysis, we then consider one of its recurring issue: dataset characterization. We then propose and evaluate such a characterization, consisting in a dissimilarity between datasets making use of a precise topological description to compare them. This dissimilarity allows a new meta-analysis approach producing recommendations of complete data analysis processes, which we then evaluate on a proof of concept. We thus detail the proposed methods of meta-analysis, and the associated process of assistance to data analysis
Le, Pévédic Brigitte. "Prédiction morphosyntaxique évolutive dans un système d'aide à la saisie de textes pour des personnes handicapées physiques - HandiAS." Nantes, 1997. http://www.theses.fr/1997NANT2058.
Full textGaüzère, Benoît. "Application des méthodes à noyaux sur graphes pour la prédiction des propriétés des molécules." Caen, 2013. http://www.theses.fr/2013CAEN2043.
Full textThis work deals with the application of graph kernel methods to the prediction of molecular properties. In this document, we first present a state of the art of graph kernels used in chemoinformatics and particurlarly those which are based on bags of patterns. Within this framework, we introduce the treelet kernel based on a set of trees which allows to encode most of the structural information encoded in molecular graphs. We also propose a combination of this kernel with multiple kernel learning methods in order to extract a subset of relevant patterns. This kernel is then extended by including cyclic information using two molecular representations defined by the relevant cycle graph and the relevant cycle hypergraph. Relevant cycle graph allows to encode the cyclic system of a molecule
Cohendet, Romain. "Prédiction computationnelle de la mémorabilité des images : vers une intégration des informations extrinsèques et émotionnelles." Thesis, Nantes, 2016. http://www.theses.fr/2016NANT4033/document.
Full textThe study of image memorability in computer science is a recent topic. First attempts were based on learning algorithms, used to infer the extent to which a picture is memorable from a set of low-level visual features. In this dissertation, we first investigate theoretical foundations of image memorability; we especially focus on the emotions the images convey, closely related to their memorability. In this light, we propose to widen the scope of image memorability prediction, to incorporate not only intrinsic, but also extrinsic image information, related to their context of presentation and to the observers. Accordingly, we build a new database for the study of image memorability; this database will be useful to test the existing models, trained on the unique database available so far. We then introduce deep learning for image memorability prediction: our model obtains the best performance to date. To improve its prediction accuracy, we try to model contextual and individual influences on image memorability. In the final part, we test the performance of computational models of visual attention, that attract growing interest for memorability prediction, for images which vary according to their degree of memorability and the emotion they convey. Finally, we present the "emotional" interactive movie, which enable us to study the links between emotion and visual attention for videos
Shi, Jie. "Influence des erreurs de transmission sur la qualité de la parole synthétique d'un vocodeur à prédiction linéaire." Toulouse 3, 1990. http://www.theses.fr/1990TOU30175.
Full textBourigault, Simon. "Apprentissage de représentations pour la prédiction de propagation d'information dans les réseaux sociaux." Thesis, Paris 6, 2016. http://www.theses.fr/2016PA066368/document.
Full textIn this thesis, we study information diffusion in online social networks. Websites like Facebook or Twitter have indeed become information medias, on which users create and share a lot of data. Most existing models of the information diffusion phenomenon relies on strong hypothesis about the structure and dynamics of diffusion. In this document, we study the problem of diffusion prediction in the context where the social graph is unknown and only user actions are observed. - We propose a learning algorithm for the independant cascades model that does not take time into account. Experimental results show that this approach obtains better results than time-based learning schemes. - We then propose several representations learning methods for this task of diffusion prediction. This let us define more compact and faster models. - Finally, we apply our representation learning approach to the source detection task, where it obtains much better results than graph-based approaches
Aich, Ali. "Reconnaissance et prédiction de situations dynamiques : application à l'assistance de personnes handicapées moteurs." Troyes, 2007. http://www.theses.fr/2007TROY0017.
Full textOur study reports on the problem of dynamic situations processing (recognition, prediction and learning) evolving in a complex and unknown environment. This study is mainly applied in the mobile robotic field. The objective is to discharge the user from the constraints of the wheelchair control by assisting him in the reproduction of routes frequently employed. The reproduction of a route already carried out requires the recognition of the new route employed. In general the recognition of this type of situations is based on two principal approaches : a recognition based on one of several models or a recognition by re-use of experiments in the field of the case-based reasoning. We have proposed during this thesis two approaches of route recognition. In the first approach, the recognition of a new route is based on Petri nets. In the second approach, the route recognition is based on similarity evaluation between the new taken route and those already carried out. Petri nets elaboration requires in general knowledge of the domain. The difficulty is that we don’t have models of routes description because on one hand we don’t use the environment maps and on the other hand it is impossible to predict from the beginning of a new displacement directions to be followed and behaviors to be carried out by the wheelchair. Hence, we have started with an initial learning. This steps corresponds to the environment exploration where a set of routes are taken for every displacement. The routes taken during the initial learning are then classified in order to compute a representative for every set of routes carried out for the same displacement. The last step of the proposed reasoning consists in modeling each each by a Petri net. The models are used during the new route recognition in order to predict the future wheelchair behaviors. In the first proposed approach, the route recognition is based on Petri nets. The new route learning is carried out only during the environment exploration. Models don’t evolve because the new not recognized routes are not taken into account. Hence, a new approach has been proposed during the second part of our work. In this approach, the new route recognition is based on a similarity evaluation between the new taken route and those already carried out. An algorithm of automatic learning of new routes is proposed so that they are used during the future reasonings. A strategy of memory maintenance is also proposed allowing the improvement of the system performances
Cumin, Julien. "Reconnaissance et prédiction d'activités dans la maison connectée." Thesis, Université Grenoble Alpes (ComUE), 2018. http://www.theses.fr/2018GREAM071/document.
Full textUnderstanding the context of a home is essential in order to provide services to occupants that fit their situations and thus fulfil their needs. One example of service that such a context-aware smart home could provide is that of a communication assistant, which can for example advise correspondents outside the home on the availability for communication of occupants. In order to implement such a service, it is indeed required that the home understands the situations of occupants, in order to derive their availability.In this thesis, we first propose a definition of context in homes. We argue that one of the primary context dimensions necessary for a system to be context-aware is the activity of occupants. As such, we then study the problem of recognizing activities, from ambient smart home sensors. We propose a new supervised place-based approach which both improves activity recognition accuracy as well as computing times compared to standard approaches.Smart home services, such as our communication assistance example, may often need to anticipate future situations. In particular, they need to anticipate future activities of occupants. Therefore, we design a new supervised activity prediction model, based on previous state-of-the-art work. We propose a number of extensions to improve prediction accuracy based on the specificities of smart home environments.Finally, we study the problem of inferring the availability of occupants for communication, in order to illustrate the feasibility of our communication assistant example. We argue that availability can be inferred from primary context dimensions such as place and activity (which can be recognized or predicted using our previous contributions), and by taking into consideration the correspondent initiating the communication as well as the modality of communication used. We discuss the impact of the activity recognition step on availability inference.We evaluate those contributions on various state-of-the-art datasets, as well as on a new dataset of activities and availabilities in homes which we constructed specifically for the purposes of this thesis: Orange4Home. Through our contributions to these 3 problems, we demonstrate the way in which an example context-aware communication assistance service can be implemented, which can advise on future availability for communication of occupants. More generally, we show how secondary context dimensions such as availability can be inferred from other context dimensions, in particular from activity. Highly accurate activity recognition and prediction are thus mandatory for a smart home to achieve context awareness
Attouche, Slimane. "Prédiction et tolérance aux fautes dans les systèmes multi-capteurs : application à la conduite automatique de véhicules en convois." Lille 1, 2002. https://pepite-depot.univ-lille.fr/RESTREINT/Th_Num/2002/50376-2002-237.pdf.
Full textKootbally, Zeïd. "Prédiction des positions de véhicules autonomes dans un environnement routier dynamique." Dijon, 2008. http://www.theses.fr/2008DIJOS064.
Full textThe goal of autonomous vehicles research is to build physical systems that can move purposefully and without human intervention in static and dynamic environments, and also in known, partially known and unknown environments. The field of autonomous vehicles is continuing to gain traction both with researchers and practitioners. Funding for research in this area has continued to grow over the past few years, and recent high profile funding opportunities have started to push theoretical research efforts into practical use. Before releasing any autonomous vehicle in the real world, it is important to model the components within a simulated environment and assess the performance of the vehicles in the virtual world. We present the PRIDE framework (Prediction In Dynamic Environments), a hierarchical multiresolutional approach for moving object prediction that incorporates multiple prediction algorithms into a single, unifying framework. PRIDE is based upon the 4D/RCS (Real-time Control System) and provides information to planners at the level of granularity that is appropriate for their planning horizon. The lower levels of the framework utilize estimation theoretic short-term predictions based upon an extended Kalman filter with an associated confidence measure. The upper levels utilize a probabilistic prédiction approach based upon situation recognition with an underlying cost model that provides predictions that incorporate environmental information and constraints. PRIDE is run in the systems' world model independently of the planner and the control system and has knowledge of the road structures via a road network database. The results of the prediction are made available to a planner to allow it to make accurate plans in dynamic environments. We have applied this approach to the visualization tool AutoSim and later on to the MOAST/USARSim framework which incorporates the physics, kinematics and dynamics of vehicles involved in traffic scenarios
Jacquemin, Ingrid. "Découverte de motifs relationnels en bioinformatique : application à la prédiction de ponts disulfures." Phd thesis, Université Rennes 1, 2005. http://tel.archives-ouvertes.fr/tel-00185499.
Full textCette thèse propose l'exploration de deux nouvelles pistes pour progresser dans la résolution de prédiction de ponts disulfures dans les protéines. Cette liaison covalente stabilise et contraint fortement la conformation spatiale de la protéine et la connaissance des positions où elle intervient peut réduire considérablement la complexité du problème de la prédiction de la structure 3D. Pour cela, nous utilisons dans un premier temps, l'inférence grammaticale et plus particulièrement les langages de contrôle introduit par Y. Takada, puis dans un deuxième temps, la programmation logique inductive.
Diverses expériences visent à confronter un cadre théorique d'apprentissage et des algorithmes généraux d'inférence grammaticale régulière à une application pratique de prédiction d'appariements spécifiques au sein d'une séquence protéique. D'autres expérimentations montrent que la programmation logique inductive donne de bons résultats sur la prédiction de l'état oxydé des cystéines en inférant des règles interprétables par les biologistes. Nous proposons un algorithme d'induction heuristique dont l'idée est d'effectuer plusieurs phases d'apprentissage en tenant compte des résultats obtenus aux phases précédentes permettant ainsi de diminuer considérablement la combinatoire dans les espaces d'hypothèses logiques en construisant des règles de plus en plus discriminantes.
Renaud, Jeremy. "Amélioration de la prédiction des commandes des pharmacies auprès de la CERP RRM." Electronic Thesis or Diss., Bourgogne Franche-Comté, 2024. http://www.theses.fr/2024UBFCD010.
Full textThe CERP Rhin Rhone Mediterranée (CERP RRM) is a wholesale distributor responsible for ensuring pharmacies' supply. Despite recent advancements in hospital logistics, the pharmaceutical sector notably lacks decision support tools. The thesis aims to establish a predictive system for all CERP clients to forecast orders with the highest possible accuracy. The data primarily consists of time series.Initially, the thesis focused on conducting a state-of-the-art review of time series prediction technologies, as well as implementing AI systems in industrial sectors related to wholesale distribution professions. The main contribution of this thesis was to enhance CERP RRM predictions at multiple levels using machine learning techniques. Our results demonstrate an improvement in predictions compared to the current method. The second contribution was to propose a new method based on sales curve analysis to group products together. This method was developed to address the issue of grouping parapharmacy products within CERP RRM. The final contribution of this thesis is a comparative study of different natural language processing models implemented in a conversational assistant for the technical service of a pharmacy management software. This solution has shown promising results, approaching those of an expert human
Frigui, Nejm Eddine. "Maintenance automatique du réseau programmable d'accès optique de très haut débit." Thesis, Ecole nationale supérieure Mines-Télécom Atlantique Bretagne Pays de la Loire, 2019. http://www.theses.fr/2019IMTA0127/document.
Full textPassive Optical Network (PON) representing one of the most attractive FTTH access network solutions, have been widely deployed for several years thanks to their ability to offer high speed services. However, due to the dynamicity of users traffic patterns, PONs need to rely on an efficient upstream bandwidth allocation mechanism. This mechanism is currently limited by the static nature of Service Level Agreement (SLA) parameters which can lead to an unoptimized bandwidth allocation in the network. The objective of this thesis is to propose a new management architecture for optimizing the upstream bandwidth allocation in PON while acting only on manageable parameters to allow the involvement of self-decision elements into the network. To achieve this, classification techniques based on machine learning approaches are used to analyze the behavior of PON users and to specify their upstream data transmission tendency. A dynamic adjustment of some SLA parameters is then performed to maximize the overall customers’ satisfaction with the network
Nguyen, Cam Linh. "Prédiction de la réponse aux traitements in vivo de tumeurs basées sur le profil moléculaire des tumeurs par apprentissage automatique." Thesis, Aix-Marseille, 2019. http://www.theses.fr/2019AIXM0208.
Full textIn recent years, targeted drugs for the treatment of cancer have been introduced. However, a drug that works in one patient may not work in another patient. To avoid the administration of ineffective treatments, methods that predict which patients will respond to a particular drug must be developed.Unfortunately, it is not currently possible to predict the effectiveness of most anticancer drugs. Machine learning (ML) is a particularly promising approach for personalized medicine. ML is a form of artificial intelligence; it concerns the development and application of computer algorithms that improve with experience. In this case, ML algorithm will learn to distinguish between sensitive and non-sensitive tumours based on multiple genes instead of a single gene. Our study focuses on applying different approaches of ML to predict drug response of tumours to anticancer drugs and generate models which have good accuracy, as well as are biologically relevant and easy to be explained
Temanni, Mohamed Ramzi. "Combinaison de sources de données pour l'amélioration de la prédiction en apprentissage : une application à la prédiction de la perte de poids chez l'obèse à partir de données transcriptomiques et cliniques." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2009. http://tel.archives-ouvertes.fr/tel-00814513.
Full textBottau, Françoise. "Contribution à l'étude de codeurs prédictifs du signal de parole avec quantificateur vectoriel." Nice, 1988. http://www.theses.fr/1988NICE4230.
Full textDermy, Oriane. "Prédiction du mouvement humain pour la robotique collaborative : du geste accompagné au mouvement corps entier." Thesis, Université de Lorraine, 2018. http://www.theses.fr/2018LORR0227/document.
Full textThis thesis lies at the intersection between machine learning and humanoid robotics, under the theme of human-robot interaction and within the cobotics (collaborative robotics) field. It focuses on prediction for non-verbal human-robot interactions, with an emphasis on gestural interaction. The prediction of the intention, understanding, and reproduction of gestures are therefore central topics of this thesis. First, the robots learn gestures by demonstration: a user grabs its arm and makes it perform the gestures to be learned several times. The robot must then be able to reproduce these different movements while generalizing them to adapt them to the situation. To do so, using its proprioceptive sensors, it interprets the perceived signals to understand the user's movement in order to generate similar ones later on. Second, the robot learns to recognize the intention of the human partner based on the gestures that the human initiates. The robot can then perform gestures adapted to the situation and corresponding to the user’s expectations. This requires the robot to understand the user’s gestures. To this end, different perceptual modalities have been explored. Using proprioceptive sensors, the robot feels the user’s gestures through its own body: it is then a question of physical human-robot interaction. Using visual sensors, the robot interprets the movement of the user’s head. Finally, using external sensors, the robot recognizes and predicts the user’s whole body movement. In that case, the user wears sensors (in our case, a wearable motion tracking suit by XSens) that transmit his posture to the robot. In addition, the coupling of these modalities was studied. From a methodological point of view, the learning and the recognition of time series (gestures) have been central to this thesis. In that aspect, two approaches have been developed. The first is based on the statistical modeling of movement primitives (corresponding to gestures) : ProMPs. The second adds Deep Learning to the first one, by using auto-encoders in order to model whole-body gestures containing a lot of information while allowing a prediction in soft real time. Various issues were taken into account during this thesis regarding the creation and development of our methods. These issues revolve around: the prediction of trajectory durations, the reduction of the cognitive and motor load imposed on the user, the need for speed (soft real-time) and accuracy in predictions
Perrier, Michel. "Observation d'état par fonctions de prédiction : maintien de la cohérence du modèle de l'environnement d'un robot mobile." Montpellier 2, 1991. http://www.theses.fr/1991MON20115.
Full textForestier, Jean-Michel. "Etude exploratoire de la prédiction en temps réel des mouvements des navires sur la houle." Cachan, Ecole normale supérieure, 2005. http://tel.archives-ouvertes.fr/tel-00011323.
Full textThe aim of this study is to establish a model of the joint behaviour of a ship and the surrounding water allowing for the short term (10 15 s) real time prediction of the motion of the ship on a swell. The proposed approach consists of 1) establishing an autonomous evolution equation deldt = f(e) for the entire system of ship and water, and 2) observing the state variables e from physical measurements at each time. The model f is based on the hypothesis of a perfect and incompressible fluid. The potential and its time derivative on the free surface or on the hull can be used as state variables of the water. Theses variables on the free surface cari. Be observed from measurement of its height. The observability of theses variables on the hull from measurement of fluid pressure is an open question. To obtain a time independant model, a perturbation method with an order zero solution not depending on time is applied
Bellón, Molina Víctor. "Prédiction personalisée des effets secondaires indésirables de médicaments." Thesis, Paris Sciences et Lettres (ComUE), 2017. http://www.theses.fr/2017PSLEM023/document.
Full textAdverse drug reaction (ADR) is a serious concern that has important health and economical repercussions. Between 1.9%-2.3% of the hospitalized patients suffer from ADR, and the annual cost of ADR have been estimated to be of 400 million euros in Germany alone. Furthermore, ADRs can cause the withdrawal of a drug from the market, which can cause up to millions of dollars of losses to the pharmaceutical industry.Multiple studies suggest that genetic factors may play a role in the response of the patients to their treatment. This covers not only the response in terms of the intended main effect, but also % according toin terms of potential side effects. The complexity of predicting drug response suggests that machine learning could bring new tools and techniques for understanding ADR.In this doctoral thesis, we study different problems related to drug response prediction, based on the genetic characteristics of patients.We frame them through multitask machine learning frameworks, which combine all data available for related problems in order to solve them at the same time.We propose a novel model for multitask linear prediction that uses task descriptors to select relevant features and make predictions with better performance as state-of-the-art algorithms. Finally, we study strategies for increasing the stability of the selected features, in order to improve interpretability for biological applications
Lepère, Stéphane. "Contribution à la prédiction en ligne des séries temporelles : un cas d'étude à la modélisation de systèmes dynamiques." Lille 1, 2001. https://pepite-depot.univ-lille.fr/RESTREINT/Th_Num/2001/50376-2001-219.pdf.
Full textBourigault, Simon. "Apprentissage de représentations pour la prédiction de propagation d'information dans les réseaux sociaux." Electronic Thesis or Diss., Paris 6, 2016. http://www.theses.fr/2016PA066368.
Full textIn this thesis, we study information diffusion in online social networks. Websites like Facebook or Twitter have indeed become information medias, on which users create and share a lot of data. Most existing models of the information diffusion phenomenon relies on strong hypothesis about the structure and dynamics of diffusion. In this document, we study the problem of diffusion prediction in the context where the social graph is unknown and only user actions are observed. - We propose a learning algorithm for the independant cascades model that does not take time into account. Experimental results show that this approach obtains better results than time-based learning schemes. - We then propose several representations learning methods for this task of diffusion prediction. This let us define more compact and faster models. - Finally, we apply our representation learning approach to the source detection task, where it obtains much better results than graph-based approaches
Wohlfarth, Till. "Machine-learning pour la prédiction des prix dans le secteur du tourisme en ligne." Electronic Thesis or Diss., Paris, ENST, 2013. http://www.theses.fr/2013ENST0090.
Full textThe goal of this paper is to consider the design of decision-making tools in the context of varying travel prices from the customer’s perspective. Based on vast streams of heterogeneous historical data collected through the internet, we describe here two approaches to forecasting travel price changes at a given horizon, taking as input variables a list of descriptive characteristics of the flight, together with possible features of the past evolution of the related price series. Though heterogeneous in many respects ( e.g. sampling, scale), the collection of historical prices series is here represented in a unified manner, by marked point processes (MPP). State-of-the-art supervised learning algorithms, possibly combined with a preliminary clustering stage, grouping flights whose related price series exhibit similar behavior, can be next used in order to help the customer to decide when to purchase her/his ticket
Al-Kharaz, Mohammed. "Analyse multivariée des alarmes de diagnostic en vue de la prédiction de la qualité des produits." Electronic Thesis or Diss., Aix-Marseille, 2021. http://theses.univ-amu.fr.lama.univ-amu.fr/211207_ALKHARAZ_559anw633vgnlp70s324svilo_TH.pdf.
Full textThis thesis addresses the prediction of product quality and improving the performance of diagnostic alarms in a semiconductor facility. For this purpose, we exploit the alarm history collected during production. First, we propose an approach to model and estimate the degradation risk of the final product associated with each alarm triggered according to its activation behavior on all products during production. Second, using the estimated risk values for any alarm, we propose an approach to predict the final quality of the product's lot. This approach models the link between process alarm events and the final quality of product lot through machine learning techniques. We also propose a new approach based on alarm event text processing to predict the final product quality. This approach improves performance and exploits more information available in the alarm text. Finally, we propose a framework for analyzing alarm activations through performance evaluation tools and several interactive visualization techniques that are more suitable for semiconductor manufacturing. These allow us to closely monitor alarms, evaluate performance, and improve the quality of products and event data collected in history. The effectiveness of each of the above approaches is demonstrated using a real data set obtained from a semiconductor manufacturing facility
Feuilloy, Mathieu. "Étude d'algorithmes d'apprentissage artificiel pour la prédiction de la syncope chez l'homme." Phd thesis, Université d'Angers, 2009. http://tel.archives-ouvertes.fr/tel-00465008.
Full textZuo, Jingwei. "Apprentissage de représentations et prédiction pour des séries-temporelles inter-dépendantes." Electronic Thesis or Diss., université Paris-Saclay, 2022. http://www.theses.fr/2022UPASG038.
Full textTime series is a common data type that has been applied to enormous real-life applications, such as financial analysis, medical diagnosis, environmental monitoring, astronomical discovery, etc. Due to its complex structure, time series raises several challenges in their data processing and mining. The representation of time series plays a key role in data mining tasks and machine learning algorithms for time series. Yet, a few methods consider the interrelation that may exist between different time series when building the representation. Moreover, the time series mining requires considering not only the time series' characteristics in terms of data complexity but also the concrete application scenarios where the data mining task is performed to build task-specific representations.In this thesis, we will study different time series representation approaches that can be used in various time series mining tasks, while capturing the relationships among them. We focus specifically on modeling the interrelations between different time series when building the representations, which can be the temporal relationship within each data source or the inter-variable relationship between various data sources. Accordingly, we study the time series collected from various application contexts under different forms. First, considering the temporal relationship between the observations, we learn the time series in a dynamic streaming context, i.e., time series stream, for which the time series data is continuously generated from the data source. Second, for the inter-variable relationship, we study the multivariate time series (MTS) with data collected from multiple data sources. Finally, we study the MTS in the Smart City context, when each data source is given a spatial position. The MTS then becomes a geo-located time series (GTS), for which the inter-variable relationship requires more modeling efforts with the external spatial information. Therefore, for each type of time series data collected from distinct contexts, the interrelations between the time series observations are emphasized differently, on the temporal or (and) variable axis.Apart from the data complexity from the interrelations, we study various machine learning tasks on time series in order to validate the learned representations. The high-level learning tasks studied in this thesis consist of time series classification, semi-supervised time series learning, and time series forecasting. We show how the learned representations connect with different time series learning tasks under distinct application contexts. More importantly, we conduct the interdisciplinary study on time series by leveraging real-life challenges in machine learning tasks, which allows for improving the learning model's performance and applying more complex time series scenarios.Concretely, for these time series learning tasks, our main research contributions are the following: (i) we propose a dynamic time series representation learning model in the streaming context, which considers both the characteristics of time series and the challenges in data streams. We claim and demonstrate that the Shapelet, a shape-based time series feature, is the best representation in such a dynamic context; (ii) we propose a semi-supervised model for representation learning in multivariate time series (MTS). The inter-variable relationship over multiple data sources is modeled in a real-life context, where the data annotations are limited; (iii) we design a geo-located time series (GTS) representation learning model for Smart City applications. We study specifically the traffic forecasting task, with a focus on the missing-value treatment within the forecasting algorithm
Santi, Nina. "Prédiction des besoins pour la gestion de serveurs mobiles en périphérie." Electronic Thesis or Diss., Université de Lille (2022-....), 2023. http://www.theses.fr/2023ULILB050.
Full textMulti-access Edge computing is an emerging paradigm within the Internet of Things (IoT) that complements Cloud computing. This paradigm proposes the implementation of computing servers located close to users, reducing the pressure and costs of local network infrastructure. This proximity to users is giving rise to new use cases, such as the deployment of mobile servers mounted on drones or robots, offering a cheaper, more energy-efficient and flexible alternative to fixed infrastructures for one-off or exceptional events. However, this approach also raises new challenges for the deployment and allocation of resources in terms of time and space, which are often battery-dependent.In this thesis, we propose predictive tools and algorithms for making decisions about the allocation of fixed and mobile resources, in terms of both time and space, within dynamic environments. We provide rich and reproducible datasets that reflect the heterogeneity inherent in Internet of Things (IoT) applications, while exhibiting a high rate of contention and interference. To achieve this, we are using the FIT-IoT Lab, an open testbed dedicated to the IoT, and we are making all the code available in an open manner. In addition, we have developed a tool for generating IoT traces in an automated and reproducible way. We use these datasets to train machine learning algorithms based on regression techniques to evaluate their ability to predict the throughput of IoT applications. In a similar approach, we have also trained and analysed a neural network of the temporal transformer type to predict several Quality of Service (QoS) metrics. In order to take into account the mobility of resources, we are generating IoT traces integrating mobile access points embedded in TurtleBot robots. These traces, which incorporate mobility, are used to validate and test a federated learning framework based on parsimonious temporal transformers. Finally, we propose a decentralised algorithm for predicting human population density by region, based on the use of a particle filter. We test and validate this algorithm using the Webots simulator in the context of servers embedded in robots, and the ns-3 simulator for the network part
Atouati, Samed. "Du texte aux chiffres : La NLP pour la prédiction des actifs financiers." Electronic Thesis or Diss., Institut polytechnique de Paris, 2022. http://www.theses.fr/2022IPPAT026.
Full textThe goal of the thesis is to investigate whether natural language data can be useful in better understanding the relationships between the public and the companies, as well as the public and the stock market price changes. In order to do so, we investigate natural language data derived from various sources: Twitter, company filings, and Reddit. We show case promising results for some sources, while others happen to have limited use as far as stock price changes go. Although they remain relevant for understanding public’s reactions to company scandals
Wohlfarth, Till. "Machine-learning pour la prédiction des prix dans le secteur du tourisme en ligne." Thesis, Paris, ENST, 2013. http://www.theses.fr/2013ENST0090/document.
Full textThe goal of this paper is to consider the design of decision-making tools in the context of varying travel prices from the customer’s perspective. Based on vast streams of heterogeneous historical data collected through the internet, we describe here two approaches to forecasting travel price changes at a given horizon, taking as input variables a list of descriptive characteristics of the flight, together with possible features of the past evolution of the related price series. Though heterogeneous in many respects ( e.g. sampling, scale), the collection of historical prices series is here represented in a unified manner, by marked point processes (MPP). State-of-the-art supervised learning algorithms, possibly combined with a preliminary clustering stage, grouping flights whose related price series exhibit similar behavior, can be next used in order to help the customer to decide when to purchase her/his ticket
Motamedi, Nikzad. "Vers la prédiction et la compréhension des effets tribologiques sur les performances systèmes par l'intelligence artificielle." Electronic Thesis or Diss., Université de Lille (2022-....), 2023. http://www.theses.fr/2023ULILN011.
Full textThe contact between two parts, especially when it is induced as in braking systems, needs to be improved and therefore to be better understood. The phenomena involved are complex because they involve multi-scale, multi-physics concepts in a context of constant evolution. An additional difficulty is that the contact is closed, and therefore it is difficult to explicitly observe the interface phenomena that play a major role in the targeted performances (noise pollution, emission of fine particles, wear etc.). In view of the European environmental standards which will be increasingly severe, it is essential to establish new strategies to better understand the problem in its entirety. The team wishes to move towards the coupling of numerical and experimental methods. The "experiment" part is based on dedicated test benches with a rich and fine instrumentation. The "numerical" part is based on multi-scale and multi-physics modeling trying to consider tribological mechanisms within a complete system. One difficulty is to compare (realign) these two parts.Thus, the objective of this PhD thesis is to propose predictive models linking the contact interface with the complete system through artificial intelligence. In the first step, we will try to determine the natural frequencies of a pin-on-disk system by considering any surface for the interface. More precisely, this interface will present a roughness field that will be generated numerically. In the second step, an AI model is developed to predict the contact distribution during a test. This part is based on measurements of a thermocouple array embedded in the near surface friction material during the tests. In the third step, based on experimental acquisitions of the surface profile at different times, a model is proposed to determine the evolution of wear. The AI models specifically developed for these three parts use algorithms such as CNN, GAN, RNN etc. These concepts are not common in the mechanical community, they are illustrated on a simple example of behavior identification in the preamble of this manuscript. In terms of results, the obtained results are very satisfactory when compared to simulation and/or experimental data. This confirms the interest of using AI in order to pass a milestone in the prediction of models. Moreover, AI also allows the understanding and the importance of the input parameters which could be used in the medium term to optimize the system or to drive the tests
Quinson, Martin. "Découverte automatique des caractéristiques et capacités d'une plate-forme de calcul distribué." Phd thesis, Ecole normale supérieure de lyon - ENS LYON, 2003. http://tel.archives-ouvertes.fr/tel-00006169.
Full textCe document est découpé en trois parties. La première présente les difficultés spécifiques à la grille en se basant sur une sélection de projets d'infrastructures pour la grille et en détaillant les solutions proposées dans ce cadre.
La seconde partie montre comment obtenir efficacement des informations quantitatives sur les capacités de la grille et leur adéquation aux besoins des routines à ordonnancer. Après avoir détaillé les problèmes rencontrés dans ce cadre, nous explicitons notre approche, nommée macro-benchmarking. Nous présentons ensuite l'outil FAST, développé dans le cadre de cette thèse et mettant cette méthodologie en oeuvre. Nous étudions également comment cet outil est utilisé dans différents projets.
La troisième partie traite de l'obtention d'une vision plus qualitative des caractéristiques de la grille, telle que la topologie d'interconnexion des machines la constituant. Après une étude des solutions classiques du domaine, nous présentons ALNeM, notre solution de cartographie automatique ne nécessitant pas de privilège d'exécution particulier. Cet outil est basé sur l'environnement GRAS, développé dans le cadre de ces travaux pour la mise au point des constituants de la grille.
Bougrain, Laurent. "Étude de la construction par réseaux neuromimétiques de représentations interprétables : application à la prédiction dans le domaine des télécommunications." Nancy 1, 2000. http://www.theses.fr/2000NAN10241.
Full textArtificial neural networks constitute good tools for certain types of computational modelling (being potentially efficient, easy to adapt and fast). However, they are often considered difficult to interpret, and are sometimes treated as black boxes. However, whilst this complexity implies that it is difficult to understand the internal organization that develops through learning, it usually encapsulates one of the key factors for obtaining good results. First, to yield a better understanding of how artificial neural networks behave and to validate their use as knowledge discovery tools, we have examined various theoretical works in order to demonstrate the common principles underlying both certain classical artificial neural network, and statistical methods for regression and data analysis. Second, in light of these studies, we have explained the specificities of some more complex artificial neural networks, such as dynamical and modular networks, in order to exploit their respective advantages in constructing a revised model for knowledge extraction, adjusted to the complexity of the phenomena we want to model. The artificial neural networks we have combined (and the subsequent model we developed) can, starting from task data, enhance the understanding of the phenomena modelled through analysing and organising the information for the task. We demonstrate this in a practical prediction task for telecommunication, where the general domain knowledge alone is insufficient to model the phenomena satisfactorily. This leads us to conclude that the possibility for practical application of out work is broad, and that our methods can combine with those already existing in the data mining and the cognitive sciences
Gerchinovitz, Sébastien. "Prédiction de suites individuelles et cadre statistique classique : étude de quelques liens autour de la régression parcimonieuse et des techniques d'agrégation." Phd thesis, Université Paris Sud - Paris XI, 2011. http://tel.archives-ouvertes.fr/tel-00653550.
Full text