The increasing complexity of trading strategies, when combined with machine learning models, forces market surveillance corporations to develop increasingly sophisticated methods for recognizing potential misuse. One strategy is to employ traders’ weapons against themselves, namely machine learning. However, the data utilized in market surveillance is highly sensitive, what may be available for machine learning is limited. In this thesis, we examine how federated learning for time series data can be used to identify potential market abuse while maintaining client privacy and data security. We are interested in developing a time-series-specific neural network employing federated learning. We demonstrate that when this strategy is used, the performance of detecting potential market abuse is comparable to that of the standard data centralized approach. Specifically, a non-federated model, a federated model, and a federated model with extra data privacy and security protection are evaluated and compared. Each model utilize an LSTM autoencoder to identify market abuse. The results demonstrate that a federated model’s performance in detecting possible market abuse is comparable to that of a non-federated model. Moreover, a federated approach with extra data privacy and security experienced a slight performance loss but is still a competitive model in comparison to the other models. Although this approach results in increased privacy and security, there is a limit to how much privacy and security can be ensured, as excessive privacy led to extremely poor performance. Federated learning offers the ability to increase data privacy and security with little performance decrease. / Den ökande komplexiteten handelsstrategier, i kombination med maskininlärning modeller, tvingar marknadsövervakning företag att utveckla allt mer sofistikerade metoder för att identifiera potentiellt marknadsmissbruk. En strategi är att använda handlarnas vapen mot sig själva, nämligen maskininlärning. Däremot, data som används inom marknadsövervakning är mycket känslig och vad som kan finnas tillgängligt för maskininlärning är begränsat.I den här studien undersöker vi hur federerat lärande för tidsseriedata kan användas till att identifiera potentiellt marknadsmissbruk samtidigt som klienternas integritet och datasäkerhet bibehålls. Vi är intresserade av att utveckla ett tidsserie-specifikt neuralt nätverk med hjälp av federated inlärning. Vi visar att när denna strategi används är prestanda för att upptäcka potentiellt marknadsmissbruk jämförbart med det för den vanliga data-centraliserade metoden. Specifikt, en icke-federerad modell, en federerad modell och en federerad modell med extra dataintegritet och säkerhet utvärderas och jämförs. Varje modell använder en LSTM-Autoencoder för att identifiera marknadsmissbruk. Resultaten visar att en federerad modells prestanda när det gäller att upptäcka eventuellt marknadsmissbruk är jämförbar med en icke-federerad modell. Dessutom, ett federerat tillvägagångssätt med extra dataintegritet upplevde en liten prestandaförlust men är fortfarande en konkurrenskraftig modell i jämförelse med andra modeller. Även om detta tillvägagångssätt resulterar i ökad integritet och säkerhet, finns det en gräns för hur mycket som kan säkerställas. Federated learning möjliggör ökad datasekretess och säkerhet med liten prestandasänkning.

Detta arbete syftar till att undersöka effektiviteten av federated learning (FL) för bildklassificering i decentraliserade databehandlingsmiljöer. Med den ökande mängden av datagenerering från mobil- och ‘edge computing’, särskilt bilddata, så finns ett behov av att förbättra metoderna för bildklassificering. Dessa metoder bör inte bara adressera de utmaningar som ställs av traditionella centraliserade djupinlärningsmodeller, utan även värna om integriteten, minska kommunikationskostnaderna och övervinna skalbarhetshinder. Federated learning erbjuder en lovande lösning som tillhandahåller en ram för modellträning över decentraliserade noder med fokus på datasekretess. Denna studie analyserar FL Förmåga att förbättra bildklassificering med dess distinkta metoder, jämför dess prestanda med konventionella modeller, och granskar dess vidare implikationer och begränsningar i praktiska, verkliga inställningar. Resultatet av denna studie visar att med lämplig hantering av brus kan FL-modeller uppnå jämförbar noggrannhet med traditionella metoder, där datasekretessen förbättras betydelsefull. Vilket demonstrerar en potential balans mellan prestanda och skydd av integritet i decentraliserade miljöer. / This study aims to explore the effectiveness of Federated Learning (FL) in image classification across decentralized computing environments. With the increasing amount of data generated from mobile and edge computing, particularly image data, there is a need to improve image classification methods that not only address the challenges posed by traditional centralized deep learning models but also respect privacy, reduce communication costs, and overcome scalability barriers. Federated Learning is a promising solution that offers a framework for model training across decentralized nodes with a focus on data privacy. This study analyzes FL's capabilities to enhance image classification using its distinct methodologies, compares its performance with conventional models, and examines its wider implications and limitations in practical, real-world settings. The result of the study indicates that with appropriate noise management, FL models can achieve comparable accuracy to traditional approaches while significantly enhancing data privacy. which demonstrates a potential balance between performance and privacy protection in decentralized environments.

Federated approach for the distributed simulation of a network, is an alternative method that aims to combine existing simulation models and software together using a Run Time Infrastructure (RTI), rather than building the whole simulation from scratch. In this study, an approach that significantly reduces the inter-federate communication load in federated simulation of communication networks is proposed. Rather than communicating packet-level information among federates, characteristics of packet flows in individual federates are dynamically identified and communicated. Flow characterization is done with the Gaussian Mixtures Algorithm (GMA) using a Self Organizing Mixture Network (SOMN) technique. In simulations of a network partitioned into eight federates in space parallel manner, it is shown that significant speedups are achieved with the proposed approach without unduly compromising accuracy.

Cloud computing brought flexibility, scalability, and capital cost savings to the IT industry. As more companies turn to cloud solutions, securing cloud based services becomes increasingly important, because for many organizations, the final barrier to adopting cloud computing is whether it is sufficiently secure. More users rely on cloud storage as it is mainly because cloud storage is available to be used by multiple devices (e.g. smart phones, tablets, notebooks, etc.) at the same time. These services often offer adequate protection to user's private data. However, there were cases where user's private data was accessible to other users, since this data is stored in a multi-tenant environment. These incidents reduce the trust of cloud storage service providers, hence there is a need to securely migrate data from one cloud storage provider to another. This thesis proposes a design of a service for providing Security as a Service for cloud brokers in a federated cloud. This scheme allows customers to securely migrate from one provider to another. To enable the design of this scheme, possible security and privacy risks of a cloud storage service were analysed and identified. Moreover, in order to successfully protect private data, data protection requirements (for data retention, sanitization, and processing) were analysed. The proposed service scheme utilizes various encryption techniques and also includes identity and key management mechanisms, such as "federated identity management". While our proposed design meets most of the defined security and privacy requirements, it is still unknown how to properly handle data sanitization, to meet data protection requirements, and provide users data recovery capabilities (backups, versioning, etc.). / Cloud computing erbjuder flexibilitet, skalbarhet, och kapital kostnadsbesparingar till IT-industrin. Eftersom fler företag vänder sig till moln lösningar, trygga molntjänster blir allt viktigare, eftersom det för många organisationer, det slutliga hindret att anta cloud computing är om det är tillräckligt säkert. Fler användare förlita sig påmoln lagring som det är främst pågrund moln lagring är tillgängligt att användas av flera enheter (t.ex. smarta telefoner, tabletter, bärbara datorer, etc.) påsamtidigt. Dessa tjänster erbjuder ofta tillräckligt skydd för användarens privata data. Men det fanns fall där användarens privata uppgifter var tillgängliga för andra användare, eftersom denna data lagras i en flera hyresgäster miljö. Dessa händelser minskar förtroende molnleverantörer lagring tjänsteleverantörer, därför finns det ett behov av att säkert migrera data från en moln lagring till en annan. Denna avhandling föreslår en utformning av en tjänst för att erbjuda säkerhet som tjänst för molnmäklare i en federativ moln. Detta system gör det möjligt för kunderna att säkert flytta från en leverantör till en annan. För att möjliggöra utformningen av detta system, möjliga säkerhet och risker integritet av ett moln lagring tjänst har analyserats och identifierats. Dessutom att man framgångsrikt skydda privata uppgifter, dataskydd krav (för data retention, sanering och bearbetning) analyserades. Den föreslagna tjänsten systemet utnyttjar olika krypteringsteknik och även inkluderar identitet och nyckelhantering mekanismer, såsom "federerad identitetshantering". Även om vår föreslagna utformningen uppfyller de flesta av den definierade säkerhet och integritet krav, är det fortfarande okänt hur korrekt hantera data sanering, för att uppfyller kraven för dataskydd och ge användarna data recovery kapacitet (säkerhetskopior, versionshantering osv.)

With the development of intelligent services and applications enabled by Artificial Intelligence (AI), the Internet of Things (IoT) is infiltrating many aspects of our everyday lives. The usability of phones and tablets is largely increasing as the primary computing device, since the powerful sensors allow these devices to have access to an unprecedented amount of data. However, there are risks and responsibilities to collect the data in a centralized location due to privacy issues. To overcome this challenge, Federated Learning (FL) allows users to collectively taking the benefits of shared models trained from this big data, without the need to centrally store it. In this research, we present and evaluate the implementation of federated learning on Raspberry Pi boards using the FedAvg method. In addition, the compression method such as quantization and sparsification was applied to the baseline implementation to improve communication efficiency. We accomplished the implementation by comparing the baseline implementation and the compressed Federated-Averaging (FedAvg) on Raspberry Pi boards in order to achieve the smallest cost and higher accuracy to fit IoT environment. / Med utvecklingen av intelligenta tjänster och applikationer möjliggjord av AI infiltrerar IoT många aspekter av vår vardag. Användbarheten för telefoner och surfplattor ökar till stor del som den primära datorenheten, eftersom de kraftfulla sensorerna tillåter dessa enheter att få tillgång till en oöverträffad mängd data. Det finns dock risker och ansvar för att lagra data på en central plats på grund av integritetsfrågor. För att övervinna denna utmaning tillåter Federated Learning (FL) användare att kollektivt ta fördelarna av delade modeller utbildade från denna stora data utan att behöva lagra den centralt. I denna forskning presenterar och utvärderar vi implementeringen av federerat lärande på Raspberry Pi-kort med FedAVG-metoden. Dessutom hade komprimeringsmetoden som kvantisering och versifiering tillämpats på basimplementeringen för att förbättra kommunikationseffektiviteten. Vi slutför implementeringen genom att jämföra baslinjeimplementeringen och den komprimerade FedAVG på Raspberry-Pi-kort för att uppnå lägsta kostnad och högre noggrannhet för att passa IoT-miljö

In this thesis a number of models for recommender systems are explored, all using collaborative filtering to produce their recommendations. Extra focus is put on two models: Matrix Factorization, which is a linear model and Multi-Layer Perceptron, which is a non-linear model. With an additional purpose of training the models without collecting any sensitive data from the users, both models were implemented with a learning technique that does not require the server's knowledge of the users' data, called federated learning. The federated version of Matrix Factorization is already well-researched, and has proven not to protect the users' data at all; the data is derivable from the information that the users communicate to the server that is necessary for the learning of the model. However, on the federated Multi-Layer Perceptron model, no research could be found. In this thesis, such a model is therefore designed and presented. Arguments are put forth in support of the privacy preservability of the model, along with a proof of the user data not being analytically derivable for the central server.    In addition, new ways to further put the protection of the users' data on the test are discussed. All models are evaluated on two different data sets. The first data set contains data on ratings of movies and is called MovieLens 1M. The second is a data set that consists of anonymized fund transactions, provided by the Swedish bank SEB for this thesis. Test results suggest that the federated versions of the models can achieve similar recommendation performance as their non-federated counterparts.

CoordenaÃÃo de AperfeiÃoamento de NÃvel Superior / Esta dissertaÃÃo apresenta uma proposta de middleware de comunicaÃÃo baseado em Enterprise Service Bus (ESB) para sistemas federados, isto Ã, formados por sistemas de diferentes organizaÃÃes. Este trabalho nÃo aborda o problema clÃssico de sistemas federados, cujo enfoque principal à autenticaÃÃo e a seguranÃa, mas sim uma necessidade crescente de intercomunicaÃÃo de serviÃos heterogÃneos. O middleware proposto, chamado de Federated Service Bus (FSB), faz uso de ESBs internos para permitir o isolamento, aplicaÃÃo de polÃticas e roteamento de cada domÃnio que compÃe o sistema federado, visando separar interesses e evitar conflitos. Nossa proposta à modelada por redes de Petri coloridas, o que lhe atribui confiabilidade de simulaÃÃo e de validaÃÃo com base em um modelo formal matemÃtico. Assim, ganhos significativos foram obtidos na implementaÃÃo com o uso de web services e BPEL (Business Process Execution Language). A modelagem com redes de Petri coloridas nÃo sà validou o fluxo, como o documentou em detalhes e possibilitou a diminuiÃÃo no nÃmero de erros. Por fim, enquadramos o FSB em arquiteturas consolidadas com SOA (Service Oriented Achitecture), EDA (Event-Driven Architecture) e NGOSS (Next Generation Operation System and Software). / This work presents the Federated Service Bus (FSB), a communication middleware based on Enterprise Service Bus (ESB) for federated systems. We do not address the classic problem of federated systems, focused mainly on authentication and security, but a growing need for heterogeneous service intercommunication. The proposed middleware makes use of internal ESBs to allow the isolation, application of policies and routing of each domain that comprises the federal system, seeking separate interests and avoid conflicts. Our proposal is modeled using coloured Petri nets, which gives it reliability of simulation and validation based on a formal mathematical model. Thus, significant gains were achieved in the implementation with the use of web services and BPEL (Business Process Execution Language). The modeling with coloured Petri nets not only validated the flow as allowed a error reduction. Finally, the FSB is embedded with SOA (Service Oriented Achitecture), EDA (Event-Driven Architecture) and NGOSS (Next Generation Operation System and Software).

Submitted by Maria Aparecida (cidazen@gmail.com) on 2017-08-04T14:25:51Z No. of bitstreams: 1 Luiz Aurélio Batista Neto.pdf: 2595761 bytes, checksum: 07f714d6c1f7297c78081b105edc8633 (MD5) / Made available in DSpace on 2017-08-04T14:25:51Z (GMT). No. of bitstreams: 1 Luiz Aurélio Batista Neto.pdf: 2595761 bytes, checksum: 07f714d6c1f7297c78081b105edc8633 (MD5) Previous issue date: 2014-10-19 / CAPES / Cloud computing applications are vulnerable to security threats originating from the Internet, because of the resources with other users and managed by third parties sharing. The diversity of services and technologies still presents a challenge to identity integration and user data in the distributed context. To address these issues, identity management techniques, especially those using a federated approach, appear crucial to protect the information from unauthorized access and allow the exchange of resources between the different trusted parties among themselves. The objective of this work is to develop a model that allows integration between identity providers through the Security Assertion Markup Language (SAML) protocol, in order to provide access to applications in multiple domains of Cloud Computing. In this scenario, each domain users and groups services as the mechanism of representation of the user according to the identity management system used (Shibboleth or SimpleSAMLphp). The proposed model is implemented to verify its applicability. In the experiments by computer simulation, the results obtained demonstrate the feasibility of the presented model. / Aplicações de Computação em Nuvem estão vulneráveis a ameaças de segurança oriundas da Internet, por conta do compartilhamento de recursos com outros usuários e gerenciados por terceiros. A diversidade de serviços e tecnologias se apresenta ainda como desafio para integração de identidades e dados de usuários no contexto distribuído. Para lidar com essas questões, técnicas de gerenciamento de identidades, especialmente as que utilizam a abordagem federada, se mostram fundamentais para proteger as informações de acessos não autorizados e permitir o intercâmbio de recursos entre as diferentes partes confiáveis entre si. O objetivo deste trabalho é desenvolver um modelo que permita a integração entre provedores de identidades por meio do protocolo Security Assertion Markup Language (SAML), com a finalidade de prover o acesso a aplicações em múltiplos domínios de Computação em Nuvem. Neste cenário, cada domínio agrupa usuários e serviços conforme o mecanismo de representação do usuário de acordo com o sistema de gerenciamento de identidades utilizado (Shibboleth ou SimpleSAMLphp). O modelo proposto é implementado para verificar a sua aplicabilidade. Nos experimentos realizados por simulação computacional, os resultados obtidos demonstram a viabilidade do modelo apresentado.

Retailers in rural Saskatchewan are having to contend with two predominant trendsrural and retail restructuring. Decreasing rural populations, increased consumer mobility, and the chronic instability of primary industries such as agriculture and forestry continue to impact rural communities in the province. The growing presence of multinational corporations, the drive for economies of scale, and the centralization of services into larger urban centers are all influencing the retail sector, particularly in rural areas. In response to these trends, retail co-operatives operating in Northern and Central Saskatchewan have joined a larger urban-based co-op in Prince Albert to form a regional co-operative. Co-operative theory suggests this regional structure may create internal obstacles for co-ops that differ from those of private firms, as co-operatives must consider the implications of reorganization on membership structures and member relations. While most of the empirical investigation has focused on large agricultural co-operatives, less attention has been afforded to consumer co-operatives.<p>Through interviews with the delegates and managers of the Prince Albert Co-operative Association (PACA), this study examines how a multi-branch consumer co-operative has adapted to the present rural and retail milieu. It investigates the new relationships that have emerged among the key stakeholders including members, delegates, and managers as well as the new relations between the major structures, namely the branches and the central body. The research is a starting point for understanding how member and enterprise interests are mediated, communicated, and coordinated within a regional co-operative. Delegates are the focal point of the study as they play an integral role in all of these relations. The findings of the study suggest that while new relationships do form within a multi-branch system, the primary relationship between members and their local co-op branch remains relatively unaffected. Further, the study on the PACA adds to Fairtloughs (2005) work on business structural forms called triarchies. It is argued that the integration of hierarchies, heterachies and responsible autonomy in the form of a federated network reinforces the staying power of the co-op in smaller communities.

Computing infrastructures offer remote access to computing power that can be employed, e.g., to solve complex mathematical problems or to host computational services that need to be online and accessible at all times. From the perspective of the infrastructure provider, large amounts of distributed and often heterogeneous computer resources need to be united into a coherent platform that is then made accessible to and usable by potential users. Grid computing and cloud computing are two paradigms that can be used to form such unified computational infrastructures. Resources from several independent infrastructure providers can be joined to form large-scale decentralized infrastructures. The primary advantage of doing this is that it increases the scale of the available resources, making it possible to address more complex problems or to run a greater number of services on the infrastructures. In addition, there are advantages in terms of factors such as fault-tolerance and geographical dispersion. Such multi-domain infrastructures require sophisticated management processes to mitigate the complications of executing computations and services across resources from different administrative domains. This thesis contributes to the development of management processes for distributed infrastructures that are designed to support multi-domain environments. It describes investigations into how fundamental management processes such as scheduling and accounting are affected by the barriers imposed by multi-domain deployments, which include technical heterogeneity, decentralized and (domain-wise) self-centric decision making, and a lack of information on the state and availability of remote resources. Four enabling technologies or approaches are explored and developed within this work: (I) The use of explicit definitions of cloud service structure as inputs for placement and management processes to ensure that the resulting placements respect the internal relationships between different service components and any relevant constraints. (II) Technology for the runtime adaptation of Virtual Machines to enable the automatic adaptation of cloud service contexts in response to changes in their environment caused by, e.g., service migration across domains. (III) Systems for managing meta-data relating to resource usage in multi-domain grid computing and cloud computing infrastructures. (IV) A global fairshare prioritization mechanism that enables computational jobs to be consistently prioritized across a federation of several decentralized grid installations. Each of these technologies will facilitate the emergence of decentralized computational infrastructures capable of utilizing resources from diverse infrastructure providers in an automatic and seamless manner. / <p>Note that the author changed surname from Henriksson to Espling in 2011</p>