Rozprawy doktorskie na temat „Analyse des réseaux sociaux complexes”
Utwórz poprawne odniesienie w stylach APA, MLA, Chicago, Harvard i wielu innych
Sprawdź 50 najlepszych rozpraw doktorskich naukowych na temat „Analyse des réseaux sociaux complexes”.
Przycisk „Dodaj do bibliografii” jest dostępny obok każdej pracy w bibliografii. Użyj go – a my automatycznie utworzymy odniesienie bibliograficzne do wybranej pracy w stylu cytowania, którego potrzebujesz: APA, MLA, Harvard, Chicago, Vancouver itp.
Możesz również pobrać pełny tekst publikacji naukowej w formacie „.pdf” i przeczytać adnotację do pracy online, jeśli odpowiednie parametry są dostępne w metadanych.
Przeglądaj rozprawy doktorskie z różnych dziedzin i twórz odpowiednie bibliografie.
Hébert-Dufresne, Laurent. "La structure communautaire comme paradigme d'organisation des réseaux complexes". Thesis, Université Laval, 2011. http://www.theses.ulaval.ca/2011/28669/28669.pdf.
Pełny tekst źródłaDugué, Nicolas. "Analyse du capitalisme social sur Twitter". Thesis, Orléans, 2015. http://www.theses.fr/2015ORLE2081/document.
Pełny tekst źródłaBourdieu, a sociologist, defines social capital as : "The set of current or potential ressources linked to the possession of a lasting relationships network". On Twitter,the friends, followers, users mentionned and retweeted are considered as the relationships network of each user, which ressources are the chance to get relevant information, to beread, to satisfy a narcissist need, to spread information or advertisements. We observethat some Twitter users that we call social capitalists aim to maximize their follower numbers to maximize their social capital. We introduce their methods, based on mutual subscriptions and dedicated hashtags. In order to study them, we first describe a large scaledetection method based on their set of followers and followees. Then, we show with an automated Twitter account that their methods allow to gain followers and to be retweeted efficiently. Afterwards, we bring to light that social capitalists methods allows these users to occupy specific positions in the network allowing them a high visibility.Furthermore, these methods make these users influent according to the major tools. Wethus set up a classification method to detect accurately these user and produce a newinfluence score
Jelassi, Mariem. "Modélisation, simulation et analyse multi-échelle de réseaux sociaux complexes : Application à l'aide à la prévention des maladies contagieuses". Thesis, Université Grenoble Alpes (ComUE), 2017. http://www.theses.fr/2017GREAS033/document.
Pełny tekst źródłaThis thesis deals with the establishment of a theoretical framework (conceptualization and formalization) capable of describing the obesity spread within a network of individuals, in order to achieve the right prevention policies and limit the epidemic spread. To do this, I started by initiating an in-depth analysis of the different obesity determinants. Once this stage completed, I developed a network model in which the relations between the individuals, (represented by the nodes of the network) are governed by rules allowing to evaluate the presence/absence of links according to their values of influence, age of the concerned nodes and their homophilic characteristics. This model, based on the age structure and demography, is constituted by two processes: the first one describes obesity at the individual level, by using epidemiological compartments. The second one describes the inter-individual level by using an individual-based network. Later, when the model reached its asymptotic behavior, I studied the social structure obtained to locate the most important individuals to be targeted in the prevention policy. Eventually, to validate the model with data, I realized an investigation in a Tunisian college and compared the obtained results from this study with those obtained from a French college survey
Ben, Chaabene Nour El Houda. "Détection d'utilisateurs violents et de menaces dans les réseaux sociaux". Electronic Thesis or Diss., Institut polytechnique de Paris, 2022. http://www.theses.fr/2022IPPAS001.
Pełny tekst źródłaOnline social networks are an integral part of people's daily social activity. They provide platforms to connect people from all over the world and share their interests. Recent statistics indicate that 56% of the world's population use these social media. However, these network services have also had many negative impacts and the existence of phenomena of aggression and intimidation in these spaces is inevitable and must therefore be addressed. Exploring the complex structure of social networks to detect violent behavior and threats is a challenge for data mining, machine learning, and artificial intelligence. In this thesis work, we aim to propose new approaches for the detection of violent behavior in social networks. Our approaches attempt to resolve this problem for several practical reasons. First, different people have different ways of expressing the same violent behavior. It is desirable to design an approach that works for everyone because of the variety of behaviors and the various ways in which they are expressed. Second, the approaches must have a way to detect potential unseen abnormal behaviors and automatically add them to the training set. Third, the multimodality and multidimensionality of the data available on social networking sites must be taken into account for the development of data mining solutions that will be able to extract relevant information useful for the detection of violent behavior. Finally, approaches must consider the time-varying nature of networks to process new users and links and automatically update built models. In the light of this and to achieve the aforementioned objectives, the main contributions of this thesis are as follows: - The first contribution proposes a model for detecting violent behavior on Twitter. This model supports the dynamic nature of the network and is capable of extracting and analyzing heterogeneous data. - The second contribution introduces an approach for detecting atypical behaviors on a multidimensional network. This approach is based on the exploration and analysis of the relationships between the individuals present on this multidimensional social structure. - The third contribution presents a framework for identifying abnormal people. This intelligent framework is based on the exploitation of a multidimensional model which takes as input multimodal data coming from several sources, capable of automatically enriching the learning set by the violent behaviors detected and considers the dynamicity of the data in order to detect new violent behaviors that appear on the network. This thesis describes achievements combining data mining techniques with new machine learning techniques. To prove the performance of our experimental results, we sums based on real data taken from three popular social networks
Combe, David. "Détection de communautés dans les réseaux d'information utilisant liens et attributs". Phd thesis, Université Jean Monnet - Saint-Etienne, 2013. http://tel.archives-ouvertes.fr/tel-01056985.
Pełny tekst źródłaHo, Thi Kim Thoa. "Modélisation et analyse des réseaux complexes associées à des informations textuelles : les apports de la prétopologie, du topic modeling et de l’apprentissage automatique à l’étude de la dynamique des réseaux sociaux, la prédiction de liens et la diffusion des sujets". Thesis, Université Paris sciences et lettres, 2020. http://www.theses.fr/2020UPSLP047.
Pełny tekst źródłaThis thesis deals with the concept of complex network associated with textual information. We are interested in the analysis of these networks with a perspective of application to social networks. Our first contribution consisted in building an analysis model for a dynamic social network using the agent based modeling (ABM) approach, author-topic modeling (ATM), and using the mathematical framework of pretopology to represent the proximity of the subjects. Our modeling is called Textual-ABM. Our proposal has been to use author-topic modeling to estimate user interest based on text content and to use pretopology to model several relationships and to represent a set of neighborhoods that is more elaborate than a simple relationship. Our second contribution concerns the diffusion of information on a "heterogeneous" social network. We propose to extend the independent cascade epidemic diffusion model (IC) and the pretopological cascade diffusion model that we call Textual-Homo-IC and Textual-PCM respectively. For Textual-Homo-IC, the probability of infection is based on homophilia (resemblance of agents) which is obtained from the textual content using the topic modeling. For Textual-PCM, a pseudo-closure function with different strong levels is proposed to realize a more complex set of neighborhoods. In addition, we propose to use supervised learning to predict the diffusion of a topic with a combination of intrinsic or external factors. Our third contribution concerns the prediction of relationships between co-authors with the addition of a new topological feature related to geographical factors and content features using topic modelling. All this work was achieved by the design of specific algorithms and validated by experiments
Martinet, Lucie. "Réseaux dynamiques de terrain : caractérisation et propriétés de diffusion en milieu hospitalier". Thesis, Lyon, École normale supérieure, 2015. http://www.theses.fr/2015ENSL1010/document.
Pełny tekst źródłaIn this thesis, we focus on tools whose aim is to extract structural and temporal properties of dynamic networks as well as diffusion characteristics which can occur on these networks. We work on specific data, from the European MOSAR project, including the network of individuals proximity from time to time during 6 months at the Brek-sur-Mer Hospital. The studied network is notable because of its three dimensions constitution : the structural one induced by the distribution of individuals into distinct services, the functional dimension due to the partition of individual into groups of socio-professional categories and the temporal dimension.For each dimension, we used tools well known from the areas of statistical physics as well as graphs theory in order to extract information which enable to describe the network properties. These methods underline the specific structure of the contacts distribution which follows the individuals distribution into services. We also highlight strong links within specific socio-professional categories. Regarding the temporal part, we extract circadian and weekly patterns and quantify the similarities of these activities. We also notice distinct behaviour within patients and staff evolution. In addition, we present tools to compare the network activity within two given periods. To finish, we use simulations techniques to extract diffusion properties of the network to find some clues in order to establish a prevention policy
Friggeri, Adrien. "A Quantitative Theory of Social Cohesion". Phd thesis, Ecole normale supérieure de lyon - ENS LYON, 2012. http://tel.archives-ouvertes.fr/tel-00737199.
Pełny tekst źródłaÉloire, Fabien. "Les réseaux interorganisationnels dans la restauration lilloise : une approche néo-structurale du marché et des processus sociaux". Thesis, Lille 1, 2009. http://www.theses.fr/2009LIL12009/document.
Pełny tekst źródłaThe thesis we sustain is rooted in two fields of research, on the one hand economic sociology, and on the other hand social network analysis. Based on an empirical case, i.e. the market of restaurants in Lille (in the north of France), three aims are pursued. The first one is sociological: we highlight the “embeddedness metaphor” for which every society has an economy, and every economy can not grow up outside a society. The second aim is theoretical: we want to take into account the relational dimension of the economic and social activities of the restaurants’ owners. The third aim is methodological: we try to apply at the interorganizational level (where the boundaries of the studied population are initially unknown) the methodology of so called “complete networks”, which was first developed for the intra-organizational level. Our analysis focuses on two social processes fundamental to the functioning of the restaurants’ market in Lille: bounded solidarity among restaurants’ owners, and regulation by social status and social capital of restaurants’ owners. The first process is described thanks to the identification and analysis of the social niches (subgroups) which are constructed by restaurants’ owners when they exchange social resources. The second process is intended from the description of the gastronomic status competition in which restaurants’owners are involved in order to be recognized on the market
Morini, Matteo. "Tools for Understanding the Dynamics of Social Networks". Thesis, Lyon, 2017. http://www.theses.fr/2017LYSEN075/document.
Pełny tekst źródłaThis thesis provides the reader with a compendium of applications of network theory; tailor-madetools suited for the purpose have been devised and implemented in a data-driven fashion. In the first part, a novel centrality metric, aptly named “bridgeness”, is presented, based on adecomposition of the standard betweenness centrality. One component, local connectivity, roughlycorresponding to the degree of a node, is set apart from the other, which evaluates longer-rangestructural properties. Indeed, the latter provides a measure of the relevance of each node in“bridging” weakly connected parts of a network; a prominent feature of the metric is its agnosticism with regard to the eventual ground truth community structure.A second application is aimed at describing dynamic features of temporal graphs which are apparent at the mesoscopic level. The dataset of choice includes 40 years of selected scientific publications.The appearance and evolution in time of a specific field of study (“wavelets”) is captured,discriminating persistent features from transient artifacts, which result from the intrinsically noisy community detection process, independently performed on successive static snapshots. The concept of “laminar stream”, on which the “complexity score” we seek to optimize is based, is introduced.In a similar vein, a network of Japanese investors has been constructed, based on a dataset which includes (indirect) information on co-owned overseas subsidiaries. A hotly debated issue in the field of industrial economics, the Miwa-Ramseyer hypothesis, has been conclusively shown to be false, at least in its strong form
Ereteo, Guillaume. "Analyse sémantique des réseaux sociaux". Phd thesis, Telecom ParisTech, 2011. http://tel.archives-ouvertes.fr/tel-00586677.
Pełny tekst źródłaMourier, Johann. "Réseaux sociaux et comportements complexes chez les requins". Paris, EPHE, 2011. http://www.theses.fr/2011EPHEA001.
Pełny tekst źródłaPortilla, Yonathan. "Etude des Réseaux Sociaux : modélisation et analyse". Thesis, Avignon, 2019. http://www.theses.fr/2019AVIG0235.
Pełny tekst źródłaCurrently social networks focus on the sharing and exchange of opinions, videos, photos, music,news and others informations, one of its objectives is to establish direct and indirect linkswith users. Social networks also promote products, people (their political or artistic image) orinfluential brands.Social networks are changing rapidly, so we’re looking to see the evolution of these sharingtools, and see how social networks change over time.We have the opportunity to study the events that occur in social networks thanks to the amount ofdata they produce. In the current market there are tools that allow the analysis of social networks,but most tools are not free, and 100% free tools disappear over time. For this reason we decidedto produce computer tools able to extract and analyse the data of the social networks studied.This study begins with the state of the art, where we describe the context of the problem, thework that led to this study and a summary of the contributions made during the thesis that wepresent briefly in the rest of the abstract.i. First we focus on the geo-linguistic fingerprint and language evolution in Twitter. Accessto content of messages sent by a group of subscribers of a social network may be usedto identify and quantify some features of a group. The feature can represent the level ofinterest in an event or product, or the popularity of an idea, or of a musical hit, or of apolitical figure. The feature can also represent how language is used and transformed,how words are written and how new grammatical rules appear.ii. Then we study the evolution of the cultural phenomenon called meme in social networks.Memes were defined by R. Dowkins as a cultural phenomenon that spreads through nongeneticforms. We examine three of the most popular memes of the internet and examinetheir impact on society in the Mediterranean countries. We use for analysing Google Trends, Topsy (a tool to measure the popularity of words on Twitter) and YouTube toquantify the impact of memes in the Mediterranean society.iii. After that we study the YouTube recommendation graph based on measurements andstochastic tools. We confirm that recommendation lists influence the views of a video.We focus on the recommendation system that boosts the popularity of videos. We buildfirst a graph that captures the recommendation system in YouTube and we study the relationshipbetween the number of views of a video and the average number of views of avideo in its recommendation list.iv. To conclude we describe the online tools available and the tools that we developed duringthe thesis. The online tools Topsy, Trendistic and Google Trends allowed us to analyseplatforms like YouTube and Twitter. We also produced tools based on API’s: in Twitterwe used the Streaming function to download and analyse tweets, with the Topsy APIwe studied the evolution of the language and the use of words, and the YouTube’s APIsallowed us to describe the behaviour on the lists of recommendations and the popularityof videos
Heymann, Sébastien. "Analyse exploratoire de flots de liens pour la détection d'événements". Phd thesis, Université Pierre et Marie Curie - Paris VI, 2013. http://tel.archives-ouvertes.fr/tel-00994766.
Pełny tekst źródłaStoica, Beck Alina. "Analyse de la structure locale des grands réseaux sociaux". Phd thesis, Université Paris-Diderot - Paris VII, 2010. http://tel.archives-ouvertes.fr/tel-00987880.
Pełny tekst źródłaStoica, Alina-Mihaela. "Analyse de la structure locale des grands réseaux sociaux". Paris 7, 2010. http://www.theses.fr/2010PA077190.
Pełny tekst źródłaThe main goal of our research was to characterize the individuals connected in a social network by analyzing the local structure of the network. For that, we proposed a method that describes the way a node (corresponding to an individual) is embedded in the network. Our method is related to the analysis of egocentred networks in sociology and to the local approach in the study of complex networks. It can be applied to small networks, to fractions of networks and also to large networks, due to its small complexity. We applied the proposed method to two large social networks, one modeling online activity on MySpace, the other one modeling mobile phone communications. In the first case we were interested in analyzing the online popularity of artists on MySpace. In the second case, we proposed and used a method for clustering nodes that are connected in a similar way to the network. We found that the distribution of mobile phone users into clusters was correlated to other characteristics of the individuals (i. E. Communication intensity and age). Although in this thesis we applied the two methods only to social networks, they can be applied in the same way to any other graph, no matter its origin
Zaidi, Faraz. "Analyse, Structure et Organisation des Réseaux Complexes". Phd thesis, Université Sciences et Technologies - Bordeaux I, 2010. http://tel.archives-ouvertes.fr/tel-00542703.
Pełny tekst źródłaCaillaut, Gaëtan. "Apprentissage d'espaces prétopologiques pour l'extraction de connaissances structurées". Electronic Thesis or Diss., Orléans, 2019. http://www.theses.fr/2019ORLE3208.
Pełny tekst źródłaPretopology is a mathematical theory whose goal is to relax the set of axioms governing the well known topology theory. Weakening the set of axioms mainly consists in redefining the pseudo-closure operator which is idempotent in topology. The non-idempotence of the pretopological pseudo-closure operator offers an appropriate framework for the modeling of various phenomena, such as iterative processes evolving throughout time. Pretopology is the outcome of the generalisation of several concepts, amongst topology but also graph theory. This thesis is divided in four main parts. The first one is an introduction to the theoretical framework of the pretopology, as well as an overview of several applications in domains where the pretopology theory shines, such as machine learning, image processing or complex systems analysis.The second part will settle the logical modeling of pretopological spaces which allows to define pretopological spaces by a logical and multi-criteria combination. This modeling enables learning algorithms to define pretopological spaces by learning a logical formula. This part will also present an unrestricted pretopological spaces learning algorithm. Unrestricted pretopological spaces can be quite hard to manipulate, especially when the studied population has some structural properties that can be described in a more restricted space. This is why the third part is dedicated to the automatic learning of pretopological spaces of type V. These spaces are defined by a set of prefilters which impose a particular structure. The LPSMI algorithm, which is the main contribution of this work, is presented in this part. This algorithm relies on the multi-instance learning principles to accurately capture the structural properties of pretopological spaces of type V. Finally, the last part consists of multiple applications of the theoretical framework presented in this thesis. Applications to lexical taxonomies extraction, community detection and extraction of temporal relations, as part of a NLP process, will be presented in order to show the usefulness, the relevance and the flexibility of pretopological spaces learning
Tabourier, Lionel. "Méthode de comparaison des topologies de graphes complexes : applications aux réseaux sociaux". Paris 6, 2010. http://www.theses.fr/2010PA066335.
Pełny tekst źródłaJlili, Mohamed Malek. "Analyse et optimisation d'efficacité de réseaux manufacturiers complexes". Thesis, Université Laval, 2013. http://www.theses.ulaval.ca/2013/29907/29907.pdf.
Pełny tekst źródłaThis thesis focuses on the analysis and optimal design of manufacturing systems composed of unreliable machinery. The considered systems can operate in an assembly / disassembly structure. Buffer stocks are placed between the machines in order to decouple them from each other. These machines can operate in degraded mode. Each machine is represented as a system with three states: nominal operation, blackout and a degraded mode. We consider that the degraded mode affects only the nominal production rate of machines and not the quality of the parts produced. To assess the rate of production of such a manufacturing system with multi-state machine (called complex), an analytical method is first explored. This method consists on replacing each machine by an equivalent one with two states, and then applying one of the classical methods for networks with binary state machines. After discovering the lack of precision of this method, we used a simulation method based on the software Simio for the optimal design of networks with multi-state machines. In this design, it is about making a joint selection of technologies and buffer sizes between machines. The objective of the optimization is to maximize the rate of production under budget constraints. Most existing works consider the problem of allocating buffer stocks for serial lines or series-parallel when machine technologies are already chosen. Our method is developed and validated using different randomly generated instances. To do this, the developed simulation model is coupled with two optimization methods. The first method uses the OptQuest optimization tool. The second method is a new heuristic based on a genetic algorithm (GA). In each method, the optimizer uses the production rate estimation carried out by the simulation tool in its objective function. Our new method (simulation / GA) is compared to an approach coupling an analytical method to a GA in the case of binary machines. The numerical results illustrate the effectiveness of our method in terms of solution quality at the expense of the less efficient computation time.
Mezghani, Manel. "Analyse des réseaux sociaux : vers une adaptation de la navigation sociale". Thesis, Toulouse 3, 2015. http://www.theses.fr/2015TOU30127/document.
Pełny tekst źródłaThe advent of Web 2.0, user-centered, has given rise to a significant amount of information (personal, collective, shared, "loved", etc.). This information is a way to help users and guide them to the information sought. However, this quantity makes access to shared information more and more difficult, given the diversity of content that may interest the user. Disorientation of the user is one of the main problems related to social media. To overcome such problem, adaptation is a standard solution that can be applied in a social context. With the evolution of these social networks, new concepts appear such as social navigation, which is a way to navigate while being influenced by other users in the network: Another important concept is that of "tag". This term is defined as social annotations created by users and associated to resources. Navigation can be therefore carried out by both links and tags. Adapting social navigation means making it more targeted for each user according to their interests. In practice, this can be done by recommending tags to each user, so he can follow or not. To adapt the social navigation, we must ensure proper detection of the user's interests and taking into account their evolution. However, we are faced with some problems: i) the detection of interest, since they can be derived from several social resources (friends, resources, tags, etc.). Their relevance is primordial to ensure adequate adaptation result. ii) updating the user profile. Indeed, the social user, is characterized by its great social activity, and therefore its interests should reflect its "real" interest each time period in order to achieve a reliable adaptation. To solve the problems affecting the quality of adaptation of social navigation quoted above, we first proposed a method for detecting the user's interests. This proposal aims to overcome the detection of irrelevant interests issues. This approach analyzes the user tags depending on the content of their respective resources. Unlike most research, who do not consider the accuracy of tags with the contents of resource, the accuracy reflects whether the user is really interested with the content or not. This is done by querying the user's network and analysis of the user annotation behavior. The approach is based on the assumption that a user annotates the resource by tags reflecting the content of this resource better reflects its "true" interests. Following the proposal of the interests of detection approach, we conducted second, the treatment of the problem of updating these interests. We were interested to the user profile enrichment techniques, performed by adding interests deemed relevant at a given time. The enrichment in a social context is performed according to social information such as neighbours who share the user behaviors in common, according to the user annotation behavior, and according to the metadata annotated resources. The choice of such information shall follow the study of their influence on the changing interests of the user. The approach we used enrichment propose recommendations (tags) according to the new tags added to the user profile. Both contributions were tested on the social database Delicious. They showed a sizeable accuracy rate. They have also proven their efficiency compared to conventional methods. In addition, the rate of ambiguity associated with the tags has been greatly reduced, thanks to the implicit filtering of irrelevant tags relative to resource content
Coupechoux, Emilie. "Analyse de grands graphes aléatoires". Paris 7, 2012. http://www.theses.fr/2012PA077184.
Pełny tekst źródłaSeveral kinds of real-world networks can be represented by graphs. Since such networks are very large, their detailed topology is generally unknown, and we model them by large random graphs having the same local statistical properties as the observed networks. An example of such properties is the fact that real-world networks are often highly clustered : if two individuals have a friend in common, they are likely to also be each other's friends. Studying random graph models that are both appropriate and tractable from a mathematical point of view is challenging, that is why we consider several clustered random graph models. The spread of epidemics in random graphs can be used to model several kinds of phenomena in real-world networks, as the spread of diseases, or the diffusion of a new technology. The epidemic model we consider depends on the phenomenon we wish to represent :. An individual can contract a disease by a single contact with any of his friends (such contacts being independent),. But a new technology is likely to be adopted by an individual if many of his friends already have the technology in question. We essentially study these two cases. In each case, one wants to know if a small proportion of the population initially infected (or having the technology in question) can propagate the epidemic to a large part of the population
Jourdan, Fabien. "Visualisation d'information : dessin, indices structuraux et navigation : Applications aux réseaux biologiques et aux réseaux sociaux". Montpellier 2, 2004. http://www.theses.fr/2004MON20205.
Pełny tekst źródłaRavot, Nicolas. "Analyse et diagnostic de réseaux filaires complexes par réflectométrie". Paris 11, 2007. http://www.theses.fr/2007PA112142.
Pełny tekst źródłaThe evolution of technologies and the communication modules involve a growing complexity of the embedded systems. These systems are smarter and use more and more sensors and others components. The increase of embedded systems implies the increase of wired network that is the physical support for the data transfer and devices supply. A wired network is composed of several kinds of cables and connectors. These systems can operate in different environments and conditions that can induce failures, because of a defective cable. Nowadays, several problems begin to appear in the wired networks. A tool for diagnosing a wired network would be greatly helpful for maintenance and monitoring. The proposed solution in this thesis allows analysing and diagnosing the health of a wired network without ambiguities. We have developed a new method, called distributed reflectometry by M-sequences, which is more effective and more reliable for analysing wired networks and which considers different aspects such as integration, precision and performance. Indeed, the diagnosis function distribution in a wired network allows apprehending a complete network and guarantees a simple reflectograms analysis without incorrect interpretations. This original method, purely numerical, is an adequate solution for embedded applications
Bigeard, Elise. "Détection et analyse de la non-adhérence médicamenteuse dans les réseaux sociaux". Thesis, Lille 3, 2019. http://www.theses.fr/2019LIL3H026.
Pełny tekst źródłaDrug non-compliance refers to situations where the patient does not follow instructions from medical authorities when taking medications. Such situations include taking too much (overuse) or too little (underuse) of medications, drinking contraindicated alcohol, or making a suicide attempt using medication. According to [HAYNES 2002] increasing drug compliance may have a bigger impact on public health than any other medical improvements. However non-compliance data are difficult to obtain since non-adherent patients are unlikely to report their behaviour to their healthcare providers. This is why we use data from social media to study drug non-compliance. Our study is applied to French-speaking forums.First we collect a corpus of messages written by users from medical forums. We build vocabularies of medication and disorder names such as used by patients. We use these vocabularies to index medications and disorders in the corpus. Then we use supervised learning and information retrieval methods to detect messages talking about non-compliance. With machine learning, we obtain 0.433 F-mesure, with up to 0.421 precision or 0.610 recall. With information retrieval, we reach 0.8 precision on the first ten results.After that, we study the content of the non-compliance messages. We identify various non-compliance situations and patient's motivations. We identify 3 main motivations: self-medication, seeking an effect besides the effect the medication was prescribed for, or being in addiction or habituation situation. Self-medication is an umbrella for several situations: avoiding an adverse effect, adjusting the medication's effect, underuse a medication seen as useless, taking decisions without a doctor's advice. Non-compliance can also happen thanks to errors or carelessness, without any particular motivation.Our work provides several kinds of result: annotated corpus with non-compliance messages, classifier for the detection of non-compliance messages, typology of non-compliance situations and analysis of the causes of non-compliance
Abid, Younes. "Analyse automatisée des risques sur la vie privée dans les réseaux sociaux". Thesis, Université de Lorraine, 2018. http://www.theses.fr/2018LORR0088/document.
Pełny tekst źródłaIn this thesis we shed the light on the danger of privacy leakage on social network. We investigate privacy breaches, design attacks, show their feasibility and study their accuracies. This approach helps us to track the origin of threats and is a first step toward designing effective countermeasures. We have first introduced a subject sensitivity measure through a questionnaire survey. Then, we have designed on-line friendship and group membership link disclosure (with certainty) attacks on the largest social network “Facebook”. These attacks successfully uncover the local network of a target using only legitimate queries. We have also designed sampling techniques to rapidly collect useful data around a target. The collected data are represented by social-attribute networks and used to perform attribute inference (with uncertainty) attacks. To increase the accuracy of attacks, we have designed cleansing algorithms. These algorithms quantify the correlation between subjects, select the most relevant ones and combat data sparsity. Finally, we have used a shallow neural network to classify the data and infer the secret values of a sensitive attribute of a given target with high accuracy measured by AUC on real datasets. The proposed algorithms in this work are included in a system called SONSAI that can help end users analyzing their local network to take the hand over their privacy
Phan, Van Long Em. "Analyse asymptotique de réseaux complexes de systèmes de réaction-diffusion". Thesis, Le Havre, 2015. http://www.theses.fr/2015LEHA0012/document.
Pełny tekst źródłaThe neuron, a fundamental unit in the nervous system, is a point of interest in many scientific disciplines. Thus, there are some mathematical models that describe their behavior by ODE or PDE systems. Many of these models can then be coupled in order to study the behavior of networks, complex systems in which the properties emerge. Firstly, this work presents the main mechanisms governing the neuron behaviour in order to understand the different models. Several models are then presented, including the FitzHugh-Nagumo one, which has a interesting dynamic. The theoretical and numerical study of the asymptotic and transitory dynamics of the aforementioned model is then proposed in the second part of this thesis. From this study, the interaction networks of ODE are built by coupling previously dynamic systems. The study of identical synchronization phenomenon in these networks shows the existence of emergent properties that can be characterized by power laws. In the third part, we focus on the study of the PDE system of FHN. As the previous part, the interaction networks of PDE are studied. We have in this section a theoretical and numerical study. In the theoretical part, we show the existence of the global attractor on the space L2(Ω)nd and give the sufficient conditions for identical synchronization. In the numerical part, we illustrate the synchronization phenomenon, also the general laws of emergence such as the power laws or the patterns formation. The diffusion effect on the synchronization is studied
Lemmouchi, Slimane. "Etude de la robustesse des graphes sociaux émergents". Phd thesis, Université Claude Bernard - Lyon I, 2012. http://tel.archives-ouvertes.fr/tel-00944441.
Pełny tekst źródłaTaramasco, Toro Carla Andrea. "Impact de l'obésité sur les structures sociales et impact des structures sociales sur l'obésité". Palaiseau, Ecole polytechnique, 2011. http://pastel.archives-ouvertes.fr/pastel-00629904.
Pełny tekst źródłaI will propose a theoretical framework (conceptualization and formalization) which seeks to model obesity as a process of transformation of one's own body determined by individual (physical and psychological), inter-individual (relational, relationship between the individual and others) and socio-cultural factors (environmental, relationship between the individual and his milieu). Individual and inter-individual factors are tied to each other in a socio-cultural context whose impact is notably related to the visibility of any body being exposed on the public stage in a non-contingent way. To investigate obesity in this multifactorial manner, this paper is divided in two main parts. First, I take into account these inseparable factors to analyze the impact through time that obese individual transformation may have on the social structure. With this aim, I develop a network model in which individual interactions are in part due to homophilic selection/deselection, i. E. Preferential attachment and detachment of inter-individual links according to characteristics of the individuals involved. Homophily is here defined as the tendency of an individual to create links with other individuals sharing similar attributes with him and to cut links with other dissimilar individuals. Homophily suggests that individuals tend to interact with those who resemble them. Second, and reciprocally, I study the role which could be played by the structure of the social fabric in the increase and current development of obesity. I evaluate the impact of micro level (i. E : relations between individuals) as well as the impact of meso level (i. E : relations between districts) and between macro level (i. E :countries). This approach highlights the necessity to integrate the dynamics of each scale to better understand the evolution of the pathology. With this aim, I use two stochastic models : epidemiological compartmental model and individual centered network model, considering three influences : exogenous heterogeneous (individual- cultural), exogenous homogeneous (individual-social) and endogenous (individual-individual). All together, this investigation of obesity will allow me to investigate the social and cultural dimension involved in being and transforming one's body
Maigrot, Cédric. "Détection de fausses informations dans les réseaux sociaux". Thesis, Rennes 1, 2019. http://www.theses.fr/2019REN1S085.
Pełny tekst źródłaFalse information are multiplying and are spreading quickly on social networks. In this thesis, we analyze the publications from a multimodal point of view between the text and the associated image. Several studies were conducted during this thesis. The first compares several types of media present on social networks and aims to discriminate them automatically. The second one allows the detection and the localization of modifications in an image thanks to the comparison with an old version of this image. Finally, we focused on merged knowledge based on the predictions of other research teams to create a single system
Serrour, Belkacem. "Détection et analyse de communautés dans les réseaux". Thesis, Lyon 1, 2010. http://www.theses.fr/2010LYO10332.
Pełny tekst źródłaThe study of the sub-structure of complex networks is of major importance to relate topology and functionality. Understanding the modular units (communities) of graphs is of utmost importance to grasping knowledge about the functionality and performance of such systems. A community is defined as a group of nodes such that connections between the nodes are denser than connections with the rest of the network. Generally, the members of one community share the same interest. Many efforts have been devoted to the analysis of the modular structure of networks. The most of these works are grouped into two parts: community detection and community analysis. Community detection consists on finding communities in networks whithout knowing there size and number. While the community analysis deals the study of the structural and semantic properties of the emerged communities, and the understanding of the functionality and the performance of the network. In this thesis, we are interested on the study of the community structures in networks. We give contributions in both community analysis and community detection parts. In the community analysis part, we study the communities of communication networks and the communities in web services. On the one hand, we study the community emergence in communication networks. We propose a classification of the emerged community structures in a given network. We model the networks by graphs and we characterize them by some parameters (network size, network density, number of resources in the network, number of providers in the network, etc.). We give also a direct correlation between the network parameters and the emerged community structures. On the other hand, we study the communities in the web service logs. We aim to discover the business protocol of services (sequences of messages exchanged between the service and a client to achieve a given goal). We analyze the log files and we model them by graphs. In our final tree graph (message graph), the paths represent the conversations (communities). In the community detection part, the main goal of our contribution is to determine communities using as building blocks triangular motifs. We propose an approach for triangle community detection based on modularity optimization using the spectral algorithm decomposition and optimization. The resulting algorithm is able to identify efficiently the best partition in communities of triangles of any given network, optimizing their correspondent modularity function
Thovex, Christophe. "Réseaux de compétences : de l'analyse des réseaux sociaux à l'analyse prédictive de connaissances". Phd thesis, Nantes, 2012. https://archive.bu.univ-nantes.fr/pollux/show/show?id=9655d57c-574a-4377-8aa1-cc682eecb122.
Pełny tekst źródłaIn 1977, Freeman formalised generic measures of Social Networks Analysis (SNA). Then, the Web “2. 0” social networks have become global networks (e. G. , FaceBook, MSN). This thesis defines a semantic model, non probabilist and predictive, for the decisional analysis of professional and institutional social networks. The presented multidisciplinary model, in parallel to the Galam sociophysics, integrates some semantic methods of natural language processing and knowledge engineering, some measures of statistic sociology and some electrodynamic laws, applied to the economic performance and social climate optimisation. It has been developped and experimented in line with the Socioprise project, funded by the French State Secretariat for the prospective and development of the digital economy
Thovex, Christophe. "Réseaux de Compétences : de l'Analyse des Réseaux Sociaux à l'Analyse Prédictive de Connaissances". Phd thesis, Université de Nantes, 2012. http://tel.archives-ouvertes.fr/tel-00697798.
Pełny tekst źródłaDeroy, Françoise Renée. "Réseaux sociaux et mobilisation de ressources, analyse sociologique du dessein de Marie de l'Incarnation". Thesis, National Library of Canada = Bibliothèque nationale du Canada, 1996. http://www.collectionscanada.ca/obj/s4/f2/dsk3/ftp04/nq21453.pdf.
Pełny tekst źródłaRifi, Mouna. "Modélisation et Analyse des Réseaux Complexes : application à la sûreté nucléaire". Thesis, Sorbonne Paris Cité, 2019. http://www.theses.fr/2019USPCD049.
Pełny tekst źródłaThis work aims to propose an adequate graph modeling approach for nuclear safety accident systems and sequences.These systems and sequences come from "Probabilistic Safety Analysis" (PSA) which is an exhaustive analysis of all possible accident scenarios, to estimate their probabilities of occurrence (by grouping them by families) and the associated consequences.Then, an analysis of the resulting networks is performed by network centrality measures. A first application consists on predicting the nuclear Risk Increase Factor, which is a PSA importance factor, using supervised learning algorithms : classification tree methods, logistic regression and ensemble learning methods, on un balanced data. Furthermore, a new synthetic centrality coefficient and a similarity measure are developed to compare the networks structures and their topological characteristics, based on their centrality vectors interdependencies. This new approach uses statistical techniques (sampling,correlation and homogeneity).The relevance and appreciation of this new measure of similarity are validated on the clustering of most popular theoretical graphs and on the prediction of the type of these graphs. Finally, an application of this approach has been conducted for the clustering of nuclear safety systems networks
Bousseau, Frédéric. "Modélisation des systèmes complexes : une approche par réseaux de Petri". Angers, 1997. http://www.theses.fr/1997ANGE0008.
Pełny tekst źródłaTackx, Raphaël. "Analyse de la structure communautaire des réseaux bipartis". Electronic Thesis or Diss., Sorbonne université, 2018. https://accesdistant.sorbonne-universite.fr/login?url=https://theses-intra.sorbonne-universite.fr/2018SORUS550.pdf.
Pełny tekst źródłaIn the real world, numerous networks appear naturally, they are everywhere, in many disciplines, for example in computer science with router networks, satellite networks, webpage networks, in biology with neural networks, in ecology with biological interaction networks, in linguistic with synonym networks, in law with legal decision networks, in economy with interbank networks, in social sciences and humanities with social networks. Generally, a network reflects the interactions between many entities of a system. These interactions have different sources, a social link or a friendship link in a social network, a cable in a router network, a chemical reaction in a protein-protein interaction network, a hyperlink in a webpage network. Furthermore, the rapid democratization of digital technology in our societies, with internet in particular, leads to create new systems which can be seen as networks. Finally, all these networks depict very specific features : they come from pratical contexts, most of the time they are big (they may be comprised of several billion of nodes and links, containing a large amount of information), they share statistical properties. In this regard, they are called real-world networks or complex networks. Nowaday, network science is a research area in its own right focusing on describing and modeling these networks in order to reveal their main features and improve our understanding of their mecanisms. Most of the works in this area use graphs formalism which provides a set of mathematical tools well suited for analyzing the topology of these networks. It exists many applications, for instance applications in spread of epidemy or computer viruses, weakness of networks in case of a breakdown, attack resilience, study for link prediction, recommandation, etc. One of the major issue is the identification of community structure. The large majority of real-world networks depicts several levels of organization in their structure. Because of there is a weak global density coupled with a strong local density, we observe that nodes are usually organized into groups, called communities, which are more internally connected than they are to the rest of the network. Moreover, these structures have a meaning in the network itself, for example communities of a social network may correspond to social groups (friends, families, etc.), communities of a protein-protein network may translate fonctions of a cell, communities may be also related to similar subjects in a webpage network [...]
Verron, Sylvain. "Diagnostic et surveillance des processus complexes par réseaux bayésiens". Phd thesis, Université d'Angers, 2007. http://tel.archives-ouvertes.fr/tel-00517101.
Pełny tekst źródłaEdouard, Amosse. "Détection et analyse d’événement dans les messages courts". Thesis, Université Côte d'Azur (ComUE), 2017. http://www.theses.fr/2017AZUR4079/document.
Pełny tekst źródłaIn the latest years, the Web has shifted from a read-only medium where most users could only consume information to an interactive medium allowing every user to create, share and comment information. The downside of social media as an information source is that often the texts are short, informal and lack contextual information. On the other hand, the Web also contains structured Knowledge Bases (KBs) that could be used to enrich the user-generated content. This dissertation investigates the potential of exploiting information from the Linked Open Data KBs to detect, classify and track events on social media, in particular Twitter. More specifically, we address 3 research questions: i) How to extract and classify messages related to events? ii) How to cluster events into fine-grained categories? and 3) Given an event, to what extent user-generated contents on social medias can contribute in the creation of a timeline of sub-events? We provide methods that rely on Linked Open Data KBs to enrich the context of social media content; we show that supervised models can achieve good generalisation capabilities through semantic linking, thus mitigating overfitting; we rely on graph theory to model the relationships between NEs and the other terms in tweets in order to cluster fine-grained events. Finally, we use in-domain ontologies and local gazetteers to identify relationships between actors involved in the same event, to create a timeline of sub-events. We show that enriching the NEs in the text with information provided by LOD KBs improves the performance of both supervised and unsupervised machine learning models
Perez, Charles. "Approche comportementale pour la sécurisation des utilisateurs de réseaux sociaux numériques mobiles". Thesis, Troyes, 2014. http://www.theses.fr/2014TROY0019/document.
Pełny tekst źródłaOur society is facing many changes in the way it communicates. The emergence of mobile terminals alongside digital social networks allows information to be shared from almost anywhere with the option of all parties being connected simultaneously. The growing use of smartphones and digital social networks in a professional context presents an opportunity, but it also exposes businesses and users to many threats, such as leakage of sensitive information, spamming, illegal access to personal data, etc.Although a significant increase in malicious activities on social platforms can be observed, currently there is no solution that ensures a completely controlled usage of digital social networks. This work aims to make a major contribution in this area through the implementation of a methodology (SPOTLIGHT) that not only uses the behaviour of profiles for evaluation purposes, but also to protect the user. This methodology relies on the assumption that smartphones, which are closely related to their owners, store and memorise traces of activity (interactions) that can be used to better protect the user online.This approach is implemented in a mobile prototype called SPOTLIGHT 1.0, which analyses traces stored in users’ smartphone to help them make the right decisions to protect their data
Dulac, Adrien. "Etude des modèles à composition mixée pour l'analyse de réseaux complexes". Thesis, Université Grenoble Alpes (ComUE), 2018. http://www.theses.fr/2018GREAM080/document.
Pełny tekst źródłaRelational data are ubiquitous in the nature and their accessibility has not ceased to increase in recent years. Those data, see as a whole, form a network, which can be represented by a data structure called a graph, where each vertex of the graph is an entity and each edge a connection between pair of vertices. Complex networks in general, such as the Web, communication networks or social network, are known to exhibit common structural properties that emerge through their graphs. In this work we emphasize two important properties called *homophilly* and *preferential attachment* that arise on most of the real-world networks. We firstly study a class of powerful *random graph models* in a Bayesian nonparametric setting, called *mixed-membership model* and we focus on showing whether the models in this class comply with the mentioned properties, after giving formal definitions in a probabilistic context of the latter. Furthermore, we empirically evaluate our findings on synthetic and real-world network datasets. Secondly, we propose a new model, which extends the former Stochastic Mixed-Membership Model, for weighted networks and we develop an efficient inference algorithm able to scale to large-scale networks
Azaza, Lobna. "Une approche pour estimer l'influence dans les réseaux complexes : application au réseau social Twitter". Thesis, Bourgogne Franche-Comté, 2019. http://www.theses.fr/2019UBFCK009/document.
Pełny tekst źródłaInfluence in complex networks and in particular Twitter has become recently a hot research topic. Detecting most influential users leads to reach a large-scale information diffusion area at low cost, something very useful in marketing or political campaigns. In this thesis, we propose a new approach that considers the several relations between users in order to assess influence in complex networks such as Twitter. We model Twitter as a multiplex heterogeneous network where users, tweets and objects are represented by nodes, and links model the different relations between them (e.g., retweets, mentions, and replies).The multiplex PageRank is applied to data from two datasets in the political field to rank candidates according to their influence. Even though the candidates' ranking reflects the reality, the multiplex PageRank scores are difficult to interpret because they are very close to each other.Thus, we want to go beyond a quantitative measure and we explore how relations between nodes in the network could reveal about the influence and propose TwitBelief, an approach to assess weighted influence of a certain node. This is based on the conjunctive combination rule from the belief functions theory that allow to combine different types of relations while expressing uncertainty about their importance weights. We experiment TwitBelief on a large amount of data gathered from Twitter during the European Elections 2014 and the French 2017 elections and deduce top influential candidates. The results show that our model is flexible enough to consider multiple interactions combination according to social scientists needs or requirements and that the numerical results of the belief theory are accurate. We also evaluate the approach over the CLEF RepLab 2014 data set and show that our approach leads to quite interesting results. We also propose two extensions of TwitBelief in order to consider the tweets content. The first is the estimation of polarized influence in Twitter network. In this extension, sentiment analysis of the tweets with the algorithm of forest decision trees allows to determine the influence polarity. The second extension is the categorization of communication styles in Twitter, it determines whether the communication style of Twitter users is informative, interactive or balanced
Diaby, Mamadou. "Méthodes pour la recommandation d’offres d’emploi dans les réseaux sociaux". Thesis, Sorbonne Paris Cité, 2015. http://www.theses.fr/2015USPCD012/document.
Pełny tekst źródłaWe are entering a new era of data mining in which the main challenge is the storing andprocessing of massive data : this is leading to a new promising research and industry field called Big data. Data are currently a new raw material coveted by businesses of all sizes and all sectors. They allow organizations to analyze, understand, model and explain phenomen a such as the behavior of their users or customers. Some companies like Google, Facebook,LinkedIn and Twitter are using user data to determine their preferences in order to make targeted advertisements to increase their revenues.This thesis has been carried out in collaboration between the laboratory L2TI andWork4, a French-American startup that offers Facebook recruitment solutions. Its main objective was the development of systems recommending relevant jobs to social network users ; the developed systems have been used to advertise job positions on social networks. After studying the literature about recommender systems, information retrieval, data mining and machine learning, we modeled social users using data they posted on their profiles, those of their social relationships together with the bag-of-words and ontology-based models. We measure the interests of users for jobs using both heuristics and models based on machine learning. The development of efficient job recommender systems involved to tackle the problem of categorization and summarization of user profiles and job descriptions. After developing job recommender systems on social networks, we developed a set of systems called Work4 Oracle that predict the audience (number of clicks) of job advertisements posted on Facebook, LinkedIn or Twitter. The analysis of the results of Work4 Oracle allows us to find and quantify factors impacting the popularity of job ads posted on social networks, these results have been compared to those of the literature of Human Resource Management. All our proposed systems deal with privacy preservation by only using the data that social network users explicitly allowed to access to ; they also deal with noisy and missing data of social network users and have been validated on real-world data provided by Work4
Karoui, Myriam. "Visibilité du capital social à travers les médias sociaux : Etudes de cas sur les dynamiques sociales de l'appropriation d'un outil d'Analyse de Réseaux Sociaux". Phd thesis, Ecole Centrale Paris, 2012. http://tel.archives-ouvertes.fr/tel-00905525.
Pełny tekst źródłaSelmane, Sid Ali. "Détection et analyse des communautés dans les réseaux sociaux : approche basée sur l'analyse formelle de concepts". Thesis, Lyon 2, 2015. http://www.theses.fr/2015LYO22004.
Pełny tekst źródłaThe study of community structure in networks became an increasingly important issue. The knowledge of core modules (communities) of networks helps us to understand how they work and behaviour, and to understand the performance of these systems. A community in a graph (network) is defined as a set of nodes that are strongly linked, but weakly linked with the rest of the graph. Members of the same community share the same interests. The originality of our research is to show that it is relevant to use formal concept analysis for community detection unlike conventional approaches using graphs. We studied several problems related to community detection in social networks : (1) the evaluation of community detection methods in the literature, (2) the detection of disjointed and overlapping communities, and (3) modelling and analysing heterogeneous social network of three-dimensional data. To assess the community detection methods proposed in the literature, we discussed this subject by studying first the state of the art that allowed us to present a classification of community detection methods by evaluating each method presented in the literature (the best known methods). For the second part, we were interested in developing a disjointed and overlapping community detection approach in homogeneous social networks from adjacency matrices (one mode data or one dimension) by exploiting techniques from formal concept analysis. We paid also a special attention to methods of modeling heterogeneous social networks. We focused in particular to three-dimensional data and proposed in this framework a modeling approach and social network analysis from three-dimensional data. This is based on a methodological framework to better understand the threedimensional aspect of this data. In addition, the analysis concerns the discovery of communities and hidden relationships between different types of individuals of these networks. The main idea lies in mining communities and rules of triadic association from these heterogeneous networks to simplify and reduce the computational complexity of this process. The results will then be used for an application recommendation of links and content to individuals in a social network
Abdaoui, Amine. "Fouille des médias sociaux français : expertise et sentiment". Thesis, Montpellier, 2016. http://www.theses.fr/2016MONTT249/document.
Pełny tekst źródłaSocial Media has changed the way we communicate between individuals, within organizations and communities. The availability of these social data opens new opportunities to understand and influence the user behavior. Therefore, Social Media Mining is experiencing a growing interest in various scientific and economic circles. In this thesis, we are specifically interested in the users of these networks whom we try to characterize in two ways: (i) their expertise and their reputations and (ii) the sentiments they express.Conventionally, social data is often mined according to its network structure. However, the textual content of the exchanged messages may reveal additional knowledge that can not be known through the analysis of the structure. Until recently, the majority of work done for the analysis of the textual content was proposed for English. The originality of this thesis is to develop methods and resources based on the textual content of the messages for French Social Media Mining.In the first axis, we initially suggest to predict the user expertise. For this, we used forums that recruit health experts to learn classification models that serve to identify messages posted by experts in any other health forum. We demonstrate that models learned on appropriate forums can be used effectively on other forums. Then, in a second step, we focus on the user reputation in these forums. The idea is to seek expressions of trust and distrust expressed in the textual content of the exchanged messages, to search the recipients of these messages and use this information to deduce users' reputation. We propose a new reputation measure that weighs the score of each response by the reputation of its author. Automatic and manual evaluations have demonstrated the effectiveness of the proposed approach.In the second axis, we focus on the extraction of sentiments (emotions and polarity). For this, we started by building a French lexicon of sentiments and emotions that we call FEEL (French Expanded Emotions Lexicon). This lexicon is built semi-automatically by translating and expanding its English counterpart NRC EmoLex. We then compare FEEL with existing French lexicons from literature on reference benchmarks. The results show that FEEL improves the classification of French texts according to their polarities and emotions. Finally, we propose to evaluate different features, methods and resources for the classification of sentiments in French. The conducted experiments have identified useful features and methods in the classification of sentiments for different types of texts. The learned systems have been particularly efficient on reference benchmarks.Generally, this work opens promising perspectives on various analytical tasks of Social Media Mining including: (i) combining multiple sources in mining Social Media users; (ii) multi-modal Social Media Mining using not just text but also image, videos, location, etc. and (iii) multilingual sentiment analysis
Sid-Ali, Ahmed. "Un processus empirique à valeurs mesures pour un système de particules en interaction appliqué aux réseaux complexes". Doctoral thesis, Université Laval, 2019. http://hdl.handle.net/20.500.11794/33730.
Pełny tekst źródłaOn propose dans cette thèse une modélisation des réseaux sociaux par des processus aléatoires à valeurs mesures. Notre démarche se base sur une approche par espace latent. Cette dernière a été utilisée dans la littérature dans le but de décrire des interactions non-observées ou latentes dans la dynamique des réseaux complexes. On caractérise les individus du réseau par des mesures de Dirac représentant leurs positions dans l’espace latent. On obtient ainsi une caractérisation du réseau en temps continu par un processus de Markov à valeurs mesures écrit comme la somme des mesures de Dirac représentant les individus. On associe au réseau trois événements aléatoires simples décrivant les arrivées et les départs d’individus suivant des horloges exponentielles en associant chaque événement à une mesure aléatoire de Poisson. Cette thèse est composée essentiellement d’un premier chapitre réservé à l’état de l’art de la littérature de la modélisation des réseaux complexes suivi d’un second chapitre introductif aux processus aléatoires à valeurs mesures. Le 3-ème et 4-ème chapitres sont constitués de deux articles co-écrits avec mon directeur de thèse, Khader Khadraoui, et sont soumis pour publication dans des journaux. Le premier article, inclus dans le chapitre 3, se compose essentiellement de la description détaillée du modèle proposé ainsi que d’une procédure de Monte Carlo permettant de générer aléatoirement des réalisations du modèle, suivi d’une analyse des propriétés théoriques du processus aléatoire à valeurs mesures sous-jacent. On explicitera notamment le générateur infinitésimal du processus de Markov qui caractérise le réseau. On s’intéressera également aux propriétés de survie et d’extinction du réseau puis on proposera une analyse asymptotique dans laquelle on démontrera, en utilisant des techniques de renormalisation, la convergence faible du processus vers une mesure déterministe solution d’un système intégro-différentiel. On terminera l’article par une étude numérique démontrant par des simulations les principales propriétés obtenues avec notre modèle. Dans le second article, inclus dans le chapitre 4, on reformule notre modèle du point de vue des graphes géométriques aléatoires. Une introduction aux graphes géométriques aléatoires est d’ailleurs proposée au chapitre 1 de cette thèse. Le but de notre démarche est d’étudier les propriétés de connectivité du réseau. Ces problématiques sont largement étudiées dans la littérature des graphes géométriques aléatoires et représentent un intérêt théorique et pratique considérable. L’idée proposée est de considérer notre modèle comme un graphe géométrique aléatoire où l’espace latent représente l’espace sous-jacent et la distribution sous-jacente est celle donnée par le processus génératif du réseau. À partir de là, la question de la connectivité du graphe se pose naturellement. En particulier, on s’intéressera à la distribution des sommets isolés, i.e. d’avoir des membres sans connexion dans le réseau. Pour cela, on pose l’hypothèse supplémentaire que chaque individu dans le graphe peut être actif ou non actif suivant une loi de Bernoulli. On démontrera alors que pour certaines valeurs du seuil de connectivité, le nombre d’individus isolés suit asymptotiquement une loi de Poisson. Aussi, la question de la détection de communautés (clustering) dans leréseau est traitée en fonction du seuil de connectivité établi. Nous terminons cette thèse par une conclusion dans laquelle on discute de la pertinence des approches proposées ainsi que des perspectives que peut offrir notre démarche. En particulier, on donne des éléments permettant de généraliser notre démarche à une classe plus large de réseaux complexes.La fin du document est consacrée aux références bibliographiques utilisées tout au long de ce travail ainsi qu’à des annexes dans lesquelles le lecteur pourra trouver des rappels utiles.
This thesis concerns the stochastic modelling of complex networks. In particular, weintroduce a new social network model based on a measure-valued stochastic processes. Individuals in the network are characterized by Dirac measures representing their positions in a virtual latent space of affinities. A continuous time network characterizationis obtained by defining an atomic measure-valued Markov process as the sum of some Dirac measures. We endow the network with a basic dynamic describing the random events of arrivals and departures following Poisson point measures. This thesis is essentially consists of a first introductory chapter to the studied problems of complex networks modelling followed by a second chapter where we present an introduction to the theory of measure-valued stochastic processes. The chapters 3 and 4 are essentially composed of two articles co-written with my thesis advisor, Khader Khadraoui and submitted to journals for publication. The first article, included in chapter 3, mainly concerns the detailed description of the proposed model and a Monte Carlo procedure allowing one to generate synthetic networks. Moreover, analysis of the principal theoretical properties of the models is proposed. In particular, the infinitesimal generator of the Markov process which characterizes the network is established. We also study the survival and extinction properties of the network. Therefore, we propose an asymptotic analysis in which we demonstrate, using a renormalization technique, the weak convergence of the network process towards a deterministic measure solution of an integro-differential system. The article is completed by a numerical study. In the second article, included in chapter 4, we reformulate our model from the point of view of random geometric graphs. An introduction to random geometric graphs framework is proposed in chapter 1. The purpose of our approach is to study the connectivity properties of the network. These issues are widely studied in the literature of random geometric graphs and represent a considerable theoretical and practical interest. The proposed idea is to consider the model as a random geometric graph where the latent space represents the underlying space and the underlying distribution is given by the generative process of the network. Therefore, the question of the connectivity of the graph arises naturally. In particular, we focus on the distribution of isolated vertices, i.e. the members with no connections in the network. To this end, we make the additional hypothesis that each individual in the network can be active or not according to a Bernoulli distribution. We then show that for some values of the connectivity threshold, the number of isolated individuals follows a Poisson distribution. In addition, the question of clustering in the network is discussed and illustrated numerically. We conclude this thesis with a conclusion and perspectives chapter in which we discuss the relevance of the proposed approaches as well as the offered perspectives.The end of the thesis is devoted to the bibliographical references used throughout this work as well as appendices in which the reader can find useful reminders.
Fernandez, Marie. "Extraction et analyse du réseau acoustique d'oiseaux sociaux". Thesis, Lyon, 2018. http://www.theses.fr/2018LYSEI030.
Pełny tekst źródłaBird populations represent a significant proportion of urban and rural biodiversity. For this purpose, the acquisition of reliable, updated and precise data on bird population can be a central factor for environmental decisions. The current classical techniques are difficult regarding human resources (banding, tracking, counting) and often invasive. Bioacoustics is a non-invasive tool for animal populations monitoring (density, migration paths...). Moreover, it has been shown in many species that the study of vocal exchanges can largely help to understand the social interactions occurring in a group. However, studying vocal exchanges can be difficult, especially when we want to assess fine scale interactions. For this reason bioacoustics have rarely been used to characterize groups’ social structure. The aim of this project was to develop techniques for the extraction of individual vocalizations in a group, and the modelling of their dynamics at a fine scale. After we developed, tested and validated our method, we used it to extract the acoustic network in a bird social species, the zebra finch, and investigate the link between acoustic and social network. Throughout different studies we showed that the group composition, more particularly its size, the presence of couples or the presence of juveniles can shape parts of the vocal dynamics. We also found that the environmental context (without any perturbation, then a context of separation for a couple, or predation in a group) can impact the vocal interactions dynamics. Thus, this project make contribution to both fundamental and applied research: in fundamental research by contributing to the study of vocal interactions dynamics to better understand the social network, and in applied research by contributing to define new standards for population monitoring
Seilles, Antoine. "Structuration de débats en ligne à l’aide d’Annotationssocio-sémantiquesVers une analyse de réseaux sociaux centrés sur l’interaction". Thesis, Montpellier 2, 2012. http://www.theses.fr/2012MON20007/document.
Pełny tekst źródłaThis pdh deals with socio-semantic annotations for e-democracy and online debates. Socio-semantic annotation are used to structurate debates. Data representation was designed to facilitate social network analysis and community detection based on opinion mining. This phd was made during the ANR project Intermed wich has to develop e-participation tools for geolocalised planning.Based on Web 2.0 trend, we define debate 2.0 concept as great scale online debates. A debate 2.0 is a debate that involves at least an important part of the inhabitants of a county and that uses web 2.0 tools. Interoperability is a main challenge of debates 2.0. If discursive annotations are a web 2.0 way of interaction between citizen, to process data from citizen participation is a complicate and expensive task. We recommand to use web semantic technology and socio-semantic annotations to represent data produced by citizen. Il will increase interoperability and easiness to create new applications and features consuming this data. We propose an annotation mecanism to structurate discussions and we have developped a platform through an agile loop with on field experiment
Hoblingre, Klein Hélène. "Réseaux sociaux professionnels : instruments d'empowerment professionnel ? : analyse de cas de consultants RH et de recruteurs sur LinkedIn". Thesis, Strasbourg, 2018. http://www.theses.fr/2018STRAG047/document.
Pełny tekst źródłaThis doctoral dissertation focuses on the use of professional network services by Human Resource staff registered on LinkedIn. The aim is to understand the new possibilities and the constraints induced by the RSNP in terms of "self-representation". The literature review shows that a LinkedIn profile can be used in two main ways : for individual and institutional uses. Moreover, it appears that the profile could potentially allow users to increase their efficiency at work by making their image known to a large number of people, to develop relationships with other professionals or develop self-confidence. An initial typology of users is proposed based on the sociology of PNS. A semiotic analysis of LinkedIn profiles and comprehensive research interviews also validate a typology of users. The results show that a cleavage between two major types of use seems to appear. The common purpose of institutional use is to be more efficient at work. At the same time, individual use mainly pursues an objective of self-mediatization