Usually, medical researchers find it cumbersome to find disease data profiles that suit their research experiments requirements. In this paper, we propose a functional architecture where medical researchers can share disease data profiles after removing patients' sensitive information. In addition, the proposed architecture is equipped with some features that facilitate collaborative discussions among researchers. Besides, some machine learning techniques are adopted for analysis and modelling of disease datasets. This way, it is expected that medical researchers can better collaborate together and perform their researches on larger patient samples obtaining more accurate and representative results. The main functionalities of the architecture are introduced. One component of the architecture, which is the evaluation engine, was implemented using Matlab showing its advantages as a tool for researchers. The case of analyzing a model for a chronic disease known as Juvenile idiopathic arthritis has been studied. Obtained results show the applicability and effectiveness of the proposed approach.

@en

While the opening of data has become a common practice for both governments and companies, many datasets are still not published since they might violate privacy regulations. The risk on privacy violations is a factor that often blocks the publication of data and results in a reserved attitude of governments and companies. Additionally, even published data, which might seem privacy compliant, can violate user privacy due to the leakage of real user identities. This paper proposes a privacy risk scoring model for open data architectures to analyse and reduce the risks associated with the opening of data. The key elements consist of a new set of open data attributes reflecting privacy risks versus benefits trades-offs. Further, these attributes are evaluated using a decision engine and a scoring matrix intro a privacy risk indicator (PRI) and a privacy risk mitigation measure (PRMM). Privacy Risk Indicator (PRI) represents the predicted value of privacy risks associated with opening such data and privacy risk mitigation measures represent the measurements need to be applied on the data to avoid the expected privacy risks. The model is exemplified through five real use cases concerning open datasets.@en

The increase in successful cyber-attacks on systems with firewalls and encryption techniques has led to the creation of Intrusion Detection Systems (IDS). Machine learning techniques are often used for these systems to predict malicious behaviour in the vague and unbalanced data. Flow-based IDS monitors only the packet headers of the network traffic and not the attached data to keep up with the growing bandwidth of networks and to maintain the privacy of the users. In this context, a multilayer perceptron approach is analysed on two different datasets and compared to a J48 Decision Tree classifier. Obtained results confirm that flow-based systems seem to be, apart from inevitable, the right way for IDS in the future and that MLP can still be useful in flow-based detection.

@en