Machine-learning (ML) enables computers to learn how to recognise patterns, make unintended decisions, or react to a dynamic environment. The effectiveness of trained machines varies because of more suitable ML algorithms or because superior training sets. Although ML algorithms are known and publicly released, training sets may not be reasonably ascertainable and, indeed, may be guarded as trade secrets. In this paper we focus our attention on ML classifiers and on the statistical information that can be unconsciously or maliciously revealed from them. We show that it is possible to infer unexpected but useful information from ML classifiers. In particular, we build a novel meta-classifier and train it to hack other classifiers, obtaining meaningful information about their training sets. Such information leakage can be exploited, for example, by a vendor to build more effective classifiers or to simply acquire trade secrets from a competitor's apparatus, potentially violating its intellectual property rights.

A novel, simple, and effective approach to modeling online user behavior extracts and analyzes digital DNA sequences from user online actions and uses Twitter as a benchmark to test the proposal. Specifically, the model obtains an incisive and compact DNA-inspired characterization of user actions. Then, standard DNA analysis techniques discriminate between genuine and spambot accounts on Twitter. An experimental campaign supports the proposal, showing its effectiveness and viability. Although Twitter spambot detection is a specific use case on a specific social media platform, the proposed methodology is platform and technology agnostic, paving the way for diverse behavioral characterization tasks.

Spambot detection in online social networks is a long-lasting challenge involving the study and design of detection techniques capable of efficiently identifying ever-evolving spammers. Recently, a new wave of social spambots has emerged, with advanced human-like characteristics that allow them to go undetected even by current state-of-the-art algorithms. In this paper, we show that efficient spambots detection can be achieved via an in-depth analysis of their collective behaviors exploiting the digital DNA technique for modeling the behaviors of social network users. Inspired by its biological counterpart, in the digital DNA representation the behavioral lifetime of a digital account is encoded in a sequence of characters. Then, we define a similarity measure for such digital DNA sequences. We build upon digital DNA and the similarity between groups of users to characterize both genuine accounts and spambots. Leveraging such a characterization, we design the Social Fingerprinting technique, which is able to discriminate among spambots and genuine accounts in both a supervised and an unsupervised fashion. We also evaluate the effectiveness of Social Fingerprinting and we compare it with three state-of-the-art detection showing the superiority of our solution. Finally, among the peculiarities of our approach is the possibility to apply off-the-shelf DNA analysis techniques to study online users behaviors and to efficiently rely on a limited number of lightweight account characteristics.

Unattended sensor networks operating in hostile environments might collect data that represents a high-value target for the adversary. The unattended sensor's inability to off-load - in real time - sensitive data to a safe external entity makes it easy for the adversary to mount a focused attack aimed at eliminating certain target data. In order to facilitate survival of this data, sensors can collectively attempt to confuse the adversary by changing its location and content, i.e., by periodically moving the data around the network and encrypting it. In this paper, we focus on data survival in unattended sensor networks faced with an adversary intent on surgically destroying data which it considers to be of high value. After motivating the problem and considering several attack flavors, we propose several simple techniques and provide their detailed evaluation.

In recent years, wireless sensor networks (WSNs) have been a very popular research topic, offering a treasure trove of systems, networking, hardware, security, and application-related problems. Much of prior research assumes that the WSN is supervised by a constantly present sink and sensors can quickly offload collected data. In this paper, we focus on unattended WSNs (UWSNs) characterized by intermittent sink presence and operation in hostile settings. Potentially lengthy intervals of sink absence offer greatly increased opportunities for attacks resulting in erasure, modification, or disclosure of sensor-collected data. This paper presents an in-depth investigation of security problems unique to UWSNs (including a new adversarial model) and proposes some simple and effective countermeasures for a certain class of attacks.