Francesco Buccafurri

Michela Fazzolari, Francesco Buccafurri, Gianluca Lax et al.

Over the last years, online reviews became very important since they can influence the purchase decision of consumers and the reputation of businesses, therefore, the practice of writing fake reviews can have severe consequences on customers and service providers. Various approaches have been proposed for detecting opinion spam in online reviews, especially based on supervised classifiers. In this contribution, we start from a set of effective features used for classifying opinion spam and we re-engineered them, by considering the Cumulative Relative Frequency Distribution of each feature. By an experimental evaluation carried out on real data from Yelp.com, we show that the use of the distributional features is able to improve the performances of classifiers.

Francesco Buccafurri, Vincenzo De Angelis, Cecilia Labrini

Digital contact tracing is one of the actions useful, in combination with other measures, to manage an epidemic diffusion of an infection disease in an after-lock-down phase. This is a very timely issue, due to the pandemic of COVID-19 we are unfortunately living. Apps for contact tracing aim to detect proximity of users and to evaluate the related risk in terms of possible contagious. Existing approaches leverage Bluetooth or GPS, or their combination, even though the prevailing approach is Bluetooth-based and relies on a decentralized model requiring the mutual exchange of ephemeral identifiers among users' smartphones. Unfortunately, a number of security and privacy concerns exist in this kind of solutions, mainly due to the exchange of identifiers, while GPS-based solutions (inherently centralized) may suffer from threats concerning massive surveillance. In this paper, we propose a solution leveraging GPS to detect proximity, and Bluetooth only to improve accuracy, without enabling exchange of identifiers. Unlike related existing solutions, no complex cryptographic mechanism is adopted, while ensuring that the server does not learn anything about locations of users.