Marco Morana

In recent years, the widespread adoption of Machine Learning (ML) at the core of complex IT systems has driven researchers to investigate the security and reliability of ML techniques. A very specific kind of threats concerns the adversary mechanisms through which an attacker could induce a classification algorithm to provide the desired output. Such strategies, known as Adversarial Machine Learning (AML), have a twofold purpose: to calculate a perturbation to be applied to the classifier’s input such that the outcome is subverted, while maintaining the underlying intent of the original data. Although any manipulation that accomplishes these goals is theoretically acceptable, in real scenarios perturbations must correspond to a set of permissible manipulations of the input, which is rarely considered in the literature. In this article, we present AdverSPAM , an AML technique designed to fool the spam account detection system of an Online Social Network (OSN). The proposed black-box evasion attack is formulated as an optimization problem that computes the adversarial sample while maintaining two important properties of the feature space, namely statistical correlation and semantic dependency . Although being demonstrated in an OSN security scenario, such an approach might be applied in other context where the aim is to perturb data described by mutually related features. Experiments conducted on a public dataset show the effectiveness of AdverSPAM compared to five state-of-the-art competitors, even in the presence of adversarial defense mechanisms.

Encryption algorithms based on block ciphers are among the most widely adopted solutions for providing information security. Over the years, a variety of methods have been proposed to evaluate the robustness of these algorithms to different types of security attacks. One of the most effective analysis techniques is differential cryptanalysis, whose aim is to study how variations in the input propagate on the output. In this work we address the modeling of differential attacks to block cipher algorithms by defining a Bayesian framework that allows a probabilistic estimation of the secret key. In order to prove the validity of the proposed approach, we present as case study a differential attack to the Data Encryption Standard (DES) which, despite being one of the methods that has been most thoroughly analyzed, is still of great interest to the scientific community since its vulnerabilities may have implications on other ciphers.

In recent years, Online Social Networks (OSNs) have radically changed the way people communicate. The most widely used platforms, such as Facebook, Youtube, and Instagram, claim more than one billion monthly active users each. Beyond these, news-oriented micro-blogging services, e.g., Twitter, are daily accessed by more than 120 million users sharing contents from all over the world. Unfortunately, legitimate users of the OSNs are mixed with malicious ones, which are interested in spreading unwanted, misleading, harmful, or discriminatory content. Spam detection in OSNs is generally approached by considering the characteristics of the account under analysis, its connection with the rest of the network, as well as data and metadata representing the content shared. However, obtaining all this information can be computationally expensive, or even unfeasible, on massive networks. Driven by these motivations, in this paper we propose SpADe, a multi-stage Spam Account Detection algorithm with reject option, whose purpose is to exploit less costly features at the early stages, while progressively extracting more complex information only for those accounts that are difficult to classify. Experimental evaluation shows the effectiveness of the proposed algorithm compared to single-stage approaches, which are much more complex in terms of features processing and classification time.

Modern smart environments pose several challenges, among which the design of intelligent algorithms aimed to assist the users. When a variety of points of interest are available, for instance, trajectory recommendations are needed to suggest users the most suitable itineraries based on their interests and contextual constraints. Unfortunately, in many cases, these interests must be explicitly requested and their lack causes the so‐called cold‐start problem. Moreover, lengthy travelling distances and excessive crowdedness of specific points of interest make itinerary planning more difficult. To address these aspects, a multi‐agent itinerary suggestion system that aims at assisting the users in an online and collaborative way is proposed. A profiling agent is responsible for the detection of groups of users whose movements are characterised by similar semantic, spatial and temporal features; then, a recommendation agent leverages contextual information and dynamically associates the current user with the trajectory clusters according to a Multi‐Armed Bandit policy. Framing the trajectory recommendation as a reinforcement learning problem permits to provide high‐quality suggestions while avoiding both cold‐start and preference elicitation issues. The effectiveness of the approach is demonstrated by some deployments in real‐life scenarios, such as smart campuses and theme parks.

The possibility of performing complex data analysis through sets of cooperating personal smart devices has recently encouraged the definition of new distributed computing paradigms. The general idea behind these approaches is to move early analysis towards theedgeof the network, while relying on other intermediate (fog) or remote (cloud) devices for computations of increasing complexity. Unfortunately, because both of their distributed nature and high degree of modularity, edge-fog-cloud computing systems are particularly prone to cyber security attacks that can be performed against every element of the infrastructure. In order to address this issue, in this paper we present SMCP, a Secure Mobile Crowdsensing Protocol for fog-based applications that exploit lightweight encryption techniques that are particularly suited for low-power mobile edge devices. In order to assess the performance of the proposed security mechanisms, we consider as case study a distributed human activity recognition scenario in which machine learning algorithms are performed by users’ personal smart devices at the edge and fog layers. The functionalities provided by SMCP have been directly compared with two state-of-the-art security protocols. Results show that our approach allows to achieve a higher degree of security while maintaining a low computational cost.

The diffusion of heterogeneous smart devices capable of capturing and analysing data about users, and/or the environment, has encouraged the growth of novel sensing methodologies. One of the most attractive scenarios in which such devices, such as smartphones, tablet computers, or activity trackers, can be exploited to infer relevant information is human activity recognition (HAR). Even though some simple HAR techniques can be directly implemented on mobile devices, in some cases, such as when complex activities need to be analysed timely, users’ smart devices can operate as part of a more complex architecture. In this article, we propose a multi-device HAR framework that exploits the fog computing paradigm to move heavy computation from the sensing layer to intermediate devices and then to the cloud. As compared to traditional cloud-based solutions, this choice allows to overcome processing and storage limitations of wearable devices while also reducing the overall bandwidth consumption. Experimental analysis aims to evaluate the performance of the entire platform in terms of accuracy of the recognition process while also highlighting the benefits it might bring in smart environments.