The mathematical operation of correlation is a very simple concept, yet has a very rich history of application in a variety of engineering fields. It is essentially nothing but a technique to measure if and to what degree two signals match each other. Since this is a very basic and universal task in a wide variety of fields such as signal processing, communications, computer vision etc., it has been an important tool. The field of pattern recognition often deals with the task of analyzing signals or useful information from signals and classifying them into classes. Very often, these classes are predetermined, and examples (templates) are available for comparison. This task naturally lends itself to the application of correlation as a tool to accomplish this goal. Thus the field of Correlation Pattern Recognition has developed over the past few decades as an important area of research. From the signal processing point of view, correlation is nothing but a filtering operation. Thus there has been a great deal of work in using concepts from filter theory to develop Correlation Filters for pattern recognition. While considerable work has been to done to develop linear correlation filters over the years, especially in the field of Automatic Target Recognition, a lot of attention has recently been paid to the development of Quadratic Correlation Filters (QCF). QCFs offer the advantages of linear filters while optimizing a bank of these simultaneously to offer much improved performance. This dissertation develops efficient QCFs that offer significant savings in storage requirements and computational complexity over existing designs. Firstly, an adaptive algorithm is presented that is able to modify the QCF coefficients as new data is observed. Secondly, a transform domain implementation of the QCF is presented that has the benefits of lower computational complexity and computational requirements while retaining excellent recognition accuracy. Finally, a two dimensional QCF is presented that holds the potential to further save on storage and computations. The techniques are developed based on the recently proposed Rayleigh Quotient Quadratic Correlation Filter (RQQCF) and simulation results are provided on synthetic and real datasets.

Les agents mobiles peuvent physiquement migrer à travers un réseau informatique dans le but d’effectuer des tâches sur des machines, ayant la capacité de leur fournir un support d’exécution. Ces agents sont considérés comme composants autonomes, une propriété qui leur permet de s'adapter à des environnements dynamiques à l'échelle d'un réseau large. Ils peuvent également échanger des informations entre eux afin de collaborer au sein de leur groupe, nous parlerons ainsi d'une communauté d'agents mobiles. Nous avons développé ce concept de communauté, en se référant aux recherches et aux études précédentes pour définir un nouveau modèle comportemental d'agent mobile. Ce modèle est utilisé pour répondre aux besoins de la surveillance logicielle. Celle ci consiste à collecter des événements à partir de plusieurs sources de données (Log, événements système…) en vue de leur analyse pour pouvoir détecter des événements anormaux. Cette démarche de surveillance s'appuie sur plusieurs types d'agents mobiles issus du même modèle. Chaque type d'agent gère un domaine fonctionnel précis. L'ensemble des ces agents constitue une communauté pouvant collaborer avec différentes autres communautés lorsqu'il existe plusieurs sites à surveiller. Les résultats de cette approche nous ont permis d'évoquer les limites liées à la taille des données collectées, ce qui nous amène à de nouvelles perspectives de recherche et à penser un agent mobile "idéal". Enfin, nous nous intéressons également à l'application de la communauté d'agent mobile pour les systèmes de détection d'intrusion et la remontée d'anomalie / Mobile agents can physically travel across a network, and perform tasks on machines, that provide agent hosting capability. These agents are autonomous; this property allows them to adapt themselves on a dynamic environment in a large network. Also, they can exchange information and data in order to collaborate within their group; in this case we can talk about community of mobile agents. We refer to previous studies and research to develop this concept of community by defining a new behavioural pattern of mobile agent. This pattern is used in monitoring software approach which consist of collecting events from various data sources (log file, OS events…) and analyse them to detect abnormal events. This approach is based on different kind of mobile agents, each kind manages some features. Whole of those mobile agents constitute a community which collaborate with other communities if there are a several sites to supervise. The results of this approach allow us to evoke some limits related to size of collected data. This limit pushes us to have a new possibility of research and probably define an ideal mobile agent. Lastly, we illustrate our mobile approach with results about intrusion detection system application to retrieve anomalies

The big data framework applied to smart meters offers an exception platform for data-driven forecasting and decision making to achieve sustainable energy efficiency. Buying-in consumer confidence through respecting occupants' energy consumption behavior and preferences towards improved participation in various energy programs is imperative but difficult to obtain. The key elements for understanding and predicting household energy consumption are activities occupants perform, appliances and the times that appliances are used, and inter-appliance dependencies. This information can be extracted from the context rich big data from smart meters, although this is challenging because: (1) it is not trivial to mine complex interdependencies between appliances from multiple concurrent data streams; (2) it is difficult to derive accurate relationships between interval based events, where multiple appliance usage persist; (3) continuous generation of the energy consumption data can trigger changes in appliance associations with time and appliances. To overcome these challenges, we propose an unsupervised progressive incremental data mining technique using frequent pattern mining (appliance-appliance associations) and cluster analysis (appliance-time associations) coupled with a Bayesian network based prediction model. The proposed technique addresses the need to analyze temporal energy consumption patterns at the appliance level, which directly reflect consumers' behaviors and provide a basis for generalizing household energy models. Extensive experiments were performed on the model with real-world datasets and strong associations were discovered. The accuracy of the proposed model for predicting multiple appliances usage outperformed support vector machine during every stage while attaining accuracy of 81.65\%, 85.90\%, 89.58\% for 25\%, 50\% and 75\% of the training dataset size respectively. Moreover, accuracy results of 81.89\%, 75.88\%, 79.23\%, 74.74\%, and 72.81\% were obtained for short-term (hours), and long-term (day, week, month, and season) energy consumption forecasts, respectively.