The Deal Analytics group of PricewaterhouseCoopers Amsterdam has requested a tool for automatising the business-to-business customer analysis. This analysis was performed manually, which left room for performance improvement. This report discusses how the a product was developed which automates the analyses After two weeks of initial research, a complete system was designed and implemented in the subsequent nine weeks. The tool consists of two distinct parts: a front-end and a back-end. The front-end allows the user to customise the analysis to its own preferences, and communicates with the back-end to efficiently perform the analysis. With the help of user evaluations, the front-end has been designed such that it is usable by any PwC employee within the Deals branch.The back-end uses data analysis techniques and machine learning to analyse customer behaviour. Strong points and growth opportunities of a company are found using techniques such as customer segmentation, regression analysis, and cross-sell analysis. The product has been tested using a variety of techniques to ensure that the software does not crash on unexpected input. The final product is evaluated based on the requirements, design goals and success criteria set at the start of the project and can be considered successful.

Generative Adversarial Networks (GANs) brought rapid developments in generating synthetic images by mimicking structures in the training data. With the list of application of GANs growing drastically, it has lately become an exciting technology to explore for designers to communicate their ideas and arts through technology and create engaging experiences for humans. Nevertheless, translating human experiences to artificial intelligence and creating visually pleasant imagery is a challenging task due to complex semantics of human perception. To address this issue, we introduce an iterative training approach in which the generated images are curated by humans and the most pleasing ones are fed back into the network to retrain. Additionally, we do a factorial analysis to investigate how the aesthetic quality and the diversity are affected by the size of training data and training length. In experiments, we validate that this method can significantly improve the aesthetic quality of generated images regardless of the dataset size and training length, however the use of smaller datasets comes with a cost of reduction in the image diversity and novelty in the output images. The aesthetic bias towards certain contexts can also deteriorate the diversity and affect the model evaluations. On the other hand, no significant relationship has been found regarding the training length, however this could possibly be due to instabilities that happen during the model convergence progress.

With the rapid growth of unstructured data across different mediums, it exposes new challenges for its analysis. To overcome this, data processing pipelines are designed with the help of different tools and technologies for the analysis of data at different stages. One of the applications which we find useful for our company is the creation of knowledge graphs for better representation and understanding of relations in the data. Knowledge graph is a structure of representing information where nodes represent the entities and edges define the relationships among them. The construction of a knowledge graph is a process of extracting meaningful information of entities and relations from unstructured textual data and storing it in a graph database. In this project, we are using Neo4j as a graph database for the efficient storage of data in the form of nodes and relations. To achieve this goal, our first research question proposes the architecture and implementation of a data processing pipeline for the construction of knowledge graphs using unstructured textual data. There are three major stages involved in our pipeline and each component is implemented in a microservice architecture. The first stage starts with the parsing of textual documents in two different formats which are PDF and PPT. In the second stage, we are applying natural language processing techniques for the extraction of meaningful information out of this raw text. In the final stage, key pieces of data are stored into a graph database(Neo4j) for the construction of knowledge graphs. We are running our pipeline on a local machine for evaluating the performance and results of each component. The core aspect of retrieving insights from this unstructured data is achieved with the use of natural language processing. In order to investigate more on this component, our second research question examines the cloud based natural language processing services from three renowned providers which are Amazon, Google and IBM. For choosing a suitable service among them, we evaluate their performance on a common data set of category Marketing from wikipedia. Based on our experimental analysis, IBM stands out among them from the perspective of the quality of output, execution time, features and cost. The adoption of a cloud based service not only leads to a faster development of business solutions but also reduces the engineering effort, its cost and maintenance of our custom implementation only with a little cost per our usage.

Cloud gaming is a new paradigm that allows users to play games in the cloud and stream them to a thin client. While there is little research about cloud gaming, containerization technologies such as Docker could provide a virtualization alternative to Virtual Machines, as these suffer from overhead and GPU pass-through constraints. With Rapture, we provide an efficient cloud gaming platform based on containerization. Stable GPU multi-tenancy is achieved by resource restrictions such as frame rate and resolution limiting. A resource-aware best-fit scheduling algorithm accomplishes workload placement in the cloud gaming cluster. The best-fit algorithm outperforms other scheduling algorithms in node utilization and preservation of Quality of Experience. Furthermore, Checkpoint and Restore technologies enable migration, maintaining a high node utilization when down-scaling services. While migration improves efficiency, Quality of Experience is affected.

Deep neural networks (DNNs) are becoming the core components of many applications running on edge devices,especially for image-based analysis, e.g., identifying objects, faces, and genders. While very successful in resource rich environments like the cloud of powerful computers, utilizing Deep Learning on edge devices for inference is not used to great extend. The resource constraints of edge devices present bottlenecks that prevent all but small networks from on-device inference. Existing solutions like compression or off-loading to the cloud sacrifice either model accuracy or inference latency. To counter the aforementioned shortcomings, Masa and EdgeCaffe are proposed as a solution. The process of DNN inference by existing frameworks is to loaded and execute a model in its entirety. As this ignores the resource limitations that are inherent to edge devices, a new framework EdgeCaffe is proposed. EdgeCaffe is able to execute models partially and thereby reducing the memory footprint of the model. Additional, EdgeCaffe allows for coordination between models when a multi-model DNN inference is executed. Masa , a memory-aware multi-DNN scheduling algorithm, featuring on modeling inter-and intra-network dependency and leveraging complimentary memory usage of each layer. Masa can consistently ensure the average response time when deterministically and stochastically executing multiple DNN-based image analysis. With extensive evaluations, Masa can consistently ensure the average response time when deterministically and stochastically executing multiple DNN-based image analyses. We extensively evaluate Masa on three configurations of Raspberry Pi and a large set of popular DNN models triggered by different arrival patterns of images. Our evaluation results show that Masa can achieve lower average response times by up to 8× and 16× for deterministic and stochastic arrivals respectively on devices with small memory, i.e., 512 MB to 1 GB, compared to the state of the art multi-DNN scheduling solutions.

Current interconnected society provides us with numerous devices communicating with one another. Exchange of data thus become an integral part in our live. Data become valuable commodity in today's setting because of their usage by individual and other interested parties. Several parties may be interested in computing a function over their data while still want to keep the information on their own data private. Prior research on computing function in privacy preserving way in the domain of smart power-grid, e-metering system, wireless sensor network, and smart phone sensing generally focus on their own application and assume a total control and the static structure of the network. Moreover, a new paradigm in the field of decentralized power-grid requires privacy preserving solution to be applicable without existence of central authority. We propose two privacy preserving data aggregation protocols in peer-to-peer network scenario where there is such central authority involved. The first protocol utilizes additive homomorphism properties of Pailier scheme and the second protocol utilizes secret sharing. Both of the protocol achieve privacy-preserving requirement of some nodes in the network, as opposed to all nodes, that are included in the aggregation set by a hop count parameter from the initiating node. This way, both of the protocols require no information of overall network structure and privacy-preserving data aggregation is achieved by being able to communicate with direct neighbors of each node in the network only.

Rabobank is currently planning a complete transition of its services to the public cloud. Currently there are around 400 DevOps teams that need to make the transition from deployment on traditional on-premise infrastructure, to deploying their products to the public cloud. The thesis project investigates what are some of the biggest challenges in adopting cloud computing technologies in an enterprise. Some of these are encouraging the staff to adopt the technology, how to embed security and compliance to the cloud computing infrastructure and which services to choose when migrating an on-premise application to Microsoft Azure Cloud. Investigation has been done on how to better encourage and inform the staff about the cloud adoption. This is achieved by improving the Cloud Awareness session (where the employees are informed for the cloud initiatives in the bank) through implementation of complete CI/CD (Continuous Integration / Continuous Deployment) pipeline of a .NET Core 2.0 application with modern HTML5 responsive layout that deploys the web application on Microsoft Azure Public Cloud and gives recommendations for testing and monitoring. Moreover, the project investigates what is needed to develop the secure and compliant feature in a huge enterprise like Rabobank with an example of development an Azure Cosmos DB feature delivered as a VSTS (Visual Studio Team Services) extension to be used by the DevOps teams in their CI/CD pipeline. Finally, an analysis is done on the performance, cost and lifecycle management of the same .NET Core 2.0 application deployed on different service offerings by Microsoft Azure Public Cloud, including Windows Server virtual machine, Azure Web App Service and Azure Kubernetes Service.

This paper aims to evaluate which automated measures of aesthetic beauty are the best predictors for human ratings of aesthetics and proposes that typicality and novelty may increase the correlation between the two. To study the correlation between these metrics, a literature study was performed to find a select amount of potentially good predictors, a pipeline was created to extract these values from each image within our dataset, a survey was conducted to vote for which images were considered most aesthetic, and finally regression analysis was performed to see which metrics offered highest correlation with the human rating data. From this we could see there were indeed a number of automated metrics that consistently scored high as predictors for the human aesthetic ratings and there was a slight improvement in the fit of the prediction model upon including novelty as a feature. However, at this moment, the improvement is not significant to conclude these features are better at predicting human ratings.