The anonymity and untraceability benefits of the Dark web account for the exponentially-increased potential of its popularity while creating a suitable womb for many illicit activities, to date. Hence, in collaboration with cybersecurity and law enforcement agencies, research has provided approaches for recognizing and classifying illicit activities with most exploiting textual dark web markets' content recognition; few such approaches use images that originated from dark web content. This paper investigates this alternative technique for recognizing illegal activities from images. In particular, we investigate label-agnostic learning techniques like One-Shot and Few-Shot learning featuring the use Siamese neural networks, a state-of-the-art approach in the field. Our solution manages to handle small-scale datasets with promising accuracy. In particular, Siamese neural networks reach 90.9% on 20-Shot experiments over a 10-class dataset; this leads us to conclude that such models are a promising and cheaper alternative to the definition of automated law-enforcing machinery over the dark web.
Untile recently crowdsourcing has been primarily conceived as an online activity to harness resources for problem solving. However the emergence of opportunistic networking (ON) has opened up crowdsourcing to the spatial domain. In this paper we bring the ON model for potential crowdsourcing in the smart city environment. We introduce cognitive features to the ON that allow users' mobile devices to become aware of the surrounding physical environment. Specifically, we exploit cognitive psychology studies on dynamic memory structures and cognitive heuristics, i.e. mental models that describe how the human brain handle decision-making amongst complex and real-time stimuli. Combined with ON, these cognitive features allow devices to act as proxies in the cyber-world of their users and exchange knowledge to deliver awareness of places in an urban environment. This is done through tags associated with locations. They represent features that are perceived by humans about a place. We consider the extent to which this knowledge becomes available to participants, using interactions with locations and other nodes. This is assessed taking into account a wide range of cognitive parameters. Outcomes are important because this functionality could support a new type of recommendation system that is independent of the traditional forms of networking.