Suchergebnisse

Part 1: Invited Paper ; International audience ; The computer security community has traditionally regarded security as a "hard" property that can be modelled and formally proven under certain simplifying assumptions. Traditional security technologies assume that computer users are either malicious, e.g. hackers or spies, or benevolent, competent and well informed about the security policies. Over the past two decades, however, computing has proliferated into all aspects of modern society and the spread of malicious software (malware) like worms, viruses and botnets have become an increasing threat. This development indicates a failure in some of the fundamental assumptions that underpin existing computer security technologies and that a new view of computer security is long overdue.In this paper, we examine traditionalmodels, policies and mechanisms of computer security in order to identify areas where the fundamental assumptions may fail. In particular, we identify areas where the "hard" security properties are based on trust in the different agents in the system and certain external agents who enforce the legislative and contractual frameworks.Trust is generally considered a "soft" security property, so building a "hard" security mechanism on trust will at most give a spongy result, unless the underlying trust assumptions are made first class citizens of the security model. In most of the work in computer security, trust assumptions are implicit and they will surely fail when the environment of the systems change, e.g. when systems are used on a global scale on the Internet. We argue that making such assumptions about trust explicit is an essential requirement for the future of system security and argue why the formalisation of computational trust is necessary when we wish to reason about system security

Part 2: Open and Smart Government ; International audience ; A design science approach is followed to develop architectural blueprints for implementing platforms to source open government applications from citizens. Zachman framework is initially used as a guide to categorize and develop the artefacts. After designing the blueprints, their usefulness is demonstrated through prototype implementation, and their potential for problem solution is evaluated from the development perspective as well as communicated to governmental peers. Contributions to research and practice include a set of blueprints covering the top levels of a platform's enterprise architecture, a reusable sourcing platform prototype, set of validated test cases for following up the implementation process, as well as success factors and lessons learned from the government perspective.

Part 1: Information Security Education ; International audience ; In this paper, we report on a pilot project which was geared to explore the possibility of enabling a pathway in Cybersecurity education between community colleges and four-year institutions where students enter the four-year institution with Junior status after graduating from the community college with a suitable Associate in Science degree. The pilot was carried out in the context of the Cybercorps®: Scholarship for Service program, also focusing on exploring whether it is possible to support students pursuing such a pathway through the Cybercorps®: Scholarship for Service program and successfully placing them in the government to complete their internship and post-graduation service requirements. The pilot included two students from a community college pursuing an Associate in Science degree in Computer Science/Engineering and then transferring to the Bachelor of Science in Cybersecurity degree program at a nearby four-year institution. This paper reviews the various components of the pilot project, discusses challenges and opportunities, and provides some suggestions for future directions from the perspective of the four-year institution.

International audience ; Digital data sharing leads to unprecedented opportunities to develop data-driven systems for supporting economic activities, the social and political life, and science. Many open-access datasets are RDF graphs, but others are CSV files, Neo4J property graphs, JSON or XML documents, etc. Potential users need to understand a dataset in order to decide if it is useful for their goal. While some datasets come with a schema and/or documentation, this is not always the case. Data summarization or schema inference tools have been proposed, specializing in XML, or JSON, or the RDF data models. In this work, we present a dataset abstraction approach, which () applies on relational, CSV, XML, JSON, RDF or Property Graph data; () computes an abstraction meant for humans (as opposed to a schema meant for a parser); () integrates Information Extraction data profiling, to also classify dataset content among a set of categories of interest to the user. Our abstractions are conceptually close to an Entity-Relationship diagram, if one allows nested and possibly heterogeneous structure within entities.

International audience ; Digital data sharing leads to unprecedented opportunities to develop data-driven systems for supporting economic activities, the social and political life, and science. Many open-access datasets are RDF graphs, but others are CSV files, Neo4J property graphs, JSON or XML documents, etc. Potential users need to understand a dataset in order to decide if it is useful for their goal. While some datasets come with a schema and/or documentation, this is not always the case. Data summarization or schema inference tools have been proposed, specializing in XML, or JSON, or the RDF data models. In this work, we present a dataset abstraction approach, which () applies on relational, CSV, XML, JSON, RDF or Property Graph data; () computes an abstraction meant for humans (as opposed to a schema meant for a parser); () integrates Information Extraction data profiling, to also classify dataset content among a set of categories of interest to the user. Our abstractions are conceptually close to an Entity-Relationship diagram, if one allows nested and possibly heterogeneous structure within entities.

International audience ; Digital data sharing leads to unprecedented opportunities to develop data-driven systems for supporting economic activities, the social and political life, and science. Many open-access datasets are RDF graphs, but others are CSV files, Neo4J property graphs, JSON or XML documents, etc. Potential users need to understand a dataset in order to decide if it is useful for their goal. While some datasets come with a schema and/or documentation, this is not always the case. Data summarization or schema inference tools have been proposed, specializing in XML, or JSON, or the RDF data models. In this work, we present a dataset abstraction approach, which () applies on relational, CSV, XML, JSON, RDF or Property Graph data; () computes an abstraction meant for humans (as opposed to a schema meant for a parser); () integrates Information Extraction data profiling, to also classify dataset content among a set of categories of interest to the user. Our abstractions are conceptually close to an Entity-Relationship diagram, if one allows nested and possibly heterogeneous structure within entities.

International audience ; Digital data sharing leads to unprecedented opportunities to develop data-driven systems for supporting economic activities, the social and political life, and science. Many open-access datasets are RDF graphs, but others are CSV files, Neo4J property graphs, JSON or XML documents, etc. Potential users need to understand a dataset in order to decide if it is useful for their goal. While some datasets come with a schema and/or documentation, this is not always the case. Data summarization or schema inference tools have been proposed, specializing in XML, or JSON, or the RDF data models. In this work, we present a dataset abstraction approach, which () applies on relational, CSV, XML, JSON, RDF or Property Graph data; () computes an abstraction meant for humans (as opposed to a schema meant for a parser); () integrates Information Extraction data profiling, to also classify dataset content among a set of categories of interest to the user. Our abstractions are conceptually close to an Entity-Relationship diagram, if one allows nested and possibly heterogeneous structure within entities.

Part 2: Regular Submissions ; International audience ; The beginning and later widespread use of the early microcomputers and home computers in the 1980s were strongly related to the emergency of the first computer games. However, this important episode has traditionally focused on accomplishments that identify key moments in the past such as identifying important firsts and successful corporate innovators. According to this, we may neglect the valuable contributions of other very different actors, such as politicians, programmers, designers, distributors, software and hardware stores, hobbyists, and fan communities. Thus, this paper is concerned with some of these myriad other subjects, including also their everyday practices. By giving voice to them, we start revealing a diverse set of activities and roles that collectively contributed to the shaping of computing technology, gaming practices and even the gaming industry in their respective local contexts.

International audience ; Agriculture resources and agriculture industrialization support information service platform is an important part of Agriculture affair support platform of west part national area of P. R. China, a key projects in the National Science & Technology Pillar Program in the Eleventh Five-year Plan Period. The system is designed as a 3 layers structure. Database support subsystem is the base layer. The middle layer contains GIS model and general reuse models; the upper layer is affair process interface and the entire manager interface. Consider of the country situation of China west, nature village is setup as a information manage unit. Famers in the village act as the registered user. Other government layer monitor the system running and distribute their index independently, system managed in the province layer. Software implement under Microsoft Windows server 2003, running as a Server/Browser style.

International audience ; Virtual Reality (VR) for Education is spreading around and more and more training centers want to provide digital training, especially using VR technol-ogies. The VR market is in full growth and many companies, startups and research teams are working on developing VR for Education on a large scale. Many domains are concerned by this learning digitalization, like science (Oliveira et al., 2019), military (Taupiac, 2018) and so on. The benefits of VR for education are numerous (Mano, 2019) and deeply investigated since the beginning of VR. There are many approaches in the process of training with VR, particularly since the commercialization of new VR headsets like HTC Vive™ and Oculus Rift™. Those headsets allow room scale tracking and are shipped with 3D space tracked controllers, granting a better user expe-rience and a higher diversity in virtual training situations. New headsets are also affordable and so not only reserved for industrials, research or military, leading to new use of VR.

International audience ; Virtual Reality (VR) for Education is spreading around and more and more training centers want to provide digital training, especially using VR technol-ogies. The VR market is in full growth and many companies, startups and research teams are working on developing VR for Education on a large scale. Many domains are concerned by this learning digitalization, like science (Oliveira et al., 2019), military (Taupiac, 2018) and so on. The benefits of VR for education are numerous (Mano, 2019) and deeply investigated since the beginning of VR. There are many approaches in the process of training with VR, particularly since the commercialization of new VR headsets like HTC Vive™ and Oculus Rift™. Those headsets allow room scale tracking and are shipped with 3D space tracked controllers, granting a better user expe-rience and a higher diversity in virtual training situations. New headsets are also affordable and so not only reserved for industrials, research or military, leading to new use of VR.

International audience ; Virtual Reality (VR) for Education is spreading around and more and more training centers want to provide digital training, especially using VR technol-ogies. The VR market is in full growth and many companies, startups and research teams are working on developing VR for Education on a large scale. Many domains are concerned by this learning digitalization, like science (Oliveira et al., 2019), military (Taupiac, 2018) and so on. The benefits of VR for education are numerous (Mano, 2019) and deeply investigated since the beginning of VR. There are many approaches in the process of training with VR, particularly since the commercialization of new VR headsets like HTC Vive™ and Oculus Rift™. Those headsets allow room scale tracking and are shipped with 3D space tracked controllers, granting a better user expe-rience and a higher diversity in virtual training situations. New headsets are also affordable and so not only reserved for industrials, research or military, leading to new use of VR.

Medicare fraud results in considerable losses for governments and insurance companies and results in higher premiums from clients. Medicare fraud costs around 13 billion euros in Europe and between 21 billion and 71 billion US dollars per year in the United States. This study aims to use artificial neural network based classifiers to predict medicare fraud. The main difficulty using machine learning techniques in fraud detection or more generally anomaly detection is that the data sets are highly imbalanced. To detect medicare frauds, we propose a multiple inputs deep neural network based classifier with a Long-short Term Memory (LSTM) autoencoder component. This architecture makes it possible to take into account many sources of data without mixing them and makes the classification task easier for the final model. The latent features extracted from the LSTM autoencoder have a strong discriminating power and separate the providers into homogeneous clusters. We use the data sets from the Centers for Medicaid and Medicare Services (CMS) of the US federal government. The CMS provides publicly available data that brings together all of the cost price requests sent by American hospitals to medicare companies. Our results show that although baseline artificial neural network give good performances, they are outperformed by our multiple inputs neural networks. We have shown that using a LSTM autoencoder to embed the provider behavior gives better results and makes the classifiers more robust to class imbalance.

Medicare fraud results in considerable losses for governments and insurance companies and results in higher premiums from clients. Medicare fraud costs around 13 billion euros in Europe and between 21 billion and 71 billion US dollars per year in the United States. This study aims to use artificial neural network based classifiers to predict medicare fraud. The main difficulty using machine learning techniques in fraud detection or more generally anomaly detection is that the data sets are highly imbalanced. To detect medicare frauds, we propose a multiple inputs deep neural network based classifier with a Long-short Term Memory (LSTM) autoencoder component. This architecture makes it possible to take into account many sources of data without mixing them and makes the classification task easier for the final model. The latent features extracted from the LSTM autoencoder have a strong discriminating power and separate the providers into homogeneous clusters. We use the data sets from the Centers for Medicaid and Medicare Services (CMS) of the US federal government. The CMS provides publicly available data that brings together all of the cost price requests sent by American hospitals to medicare companies. Our results show that although baseline artificial neural network give good performances, they are outperformed by our multiple inputs neural networks. We have shown that using a LSTM autoencoder to embed the provider behavior gives better results and makes the classifiers more robust to class imbalance.

Medicare fraud results in considerable losses for governments and insurance companies and results in higher premiums from clients. Medicare fraud costs around 13 billion euros in Europe and between 21 billion and 71 billion US dollars per year in the United States. This study aims to use artificial neural network based classifiers to predict medicare fraud. The main difficulty using machine learning techniques in fraud detection or more generally anomaly detection is that the data sets are highly imbalanced. To detect medicare frauds, we propose a multiple inputs deep neural network based classifier with a Long-short Term Memory (LSTM) autoencoder component. This architecture makes it possible to take into account many sources of data without mixing them and makes the classification task easier for the final model. The latent features extracted from the LSTM autoencoder have a strong discriminating power and separate the providers into homogeneous clusters. We use the data sets from the Centers for Medicaid and Medicare Services (CMS) of the US federal government. The CMS provides publicly available data that brings together all of the cost price requests sent by American hospitals to medicare companies. Our results show that although baseline artificial neural network give good performances, they are outperformed by our multiple inputs neural networks. We have shown that using a LSTM autoencoder to embed the provider behavior gives better results and makes the classifiers more robust to class imbalance.