Search

Traffic forecasting models rely on data that needs to be sensed, processed, and stored. This requires the deployment and maintenance of traffic sensing infrastructure, often leading to unaffordable monetary costs. The lack of sensed locations can be complemented with synthetic data simulations that further lower the economical investment needed for traffic monitoring. One of the most common data generative approaches consists of producing real-like traffic patterns, according to data distributions from analogous roads. The process of detecting roads with similar traffic is the key point of these systems. However, without collecting data at the target location no flow metrics can be employed for this similarity-based search. We present a method to discover locations among those with available traffic data by inspecting topological features. These features are extracted from domain-specific knowledge as numerical representations (embeddings) to compare different locations and eventually find roads with analogous daily traffic profiles based on the similarity between embeddings. The performance of this novel selection system is examined and compared to simpler traffic estimation approaches. After finding a similar source of data, a generative method is used to synthesize traffic profiles. Depending on the resemblance of the traffic behavior at the sensed road, the generation method can be fed with data from one road only. Several generation approaches are analyzed in terms of the precision of the synthesized samples. Above all, this work intends to stimulate further research efforts towards enhancing the quality of synthetic traffic samples and thereby, reducing the need for sensing infrastructure.

Cyber attacks have recently gained momentum in the research community as a sharply concerning phenomenon further ignited by the proliferation of social networks, which unfold a variety of ways for cybercriminals to access compromised information of their users. This paper gravitates on impersonation attacks, whose motivation may go beyond economic interests of the attacker towards getting unauthorised access to information and contacts, as often occurs between teenagers and early users of social platforms. This manuscript proposes a meta-heuristically optimised learning model as the algorithmic core of a non-intrusive detection system that relies exclusively on connection time features to detect evidences of an impersonation attack. The proposed scheme hinges on the K-Means clustering approach applied to a set of time features specially tailored to characterise the usage of users, which are weighted prior to the clustering under detection performance maximisation criteria. The obtained results shed light on the potentiality of the proposed methodology for its practical application to real social networks.