Abstract
We consider privacy of obfuscated location reports that can be correlated through time/space to estimate the real position of a user. We propose a user-centric Location Privacy Preserving Mechanism (LPPM) that protects users not only against single reports, but also over time, against continuous reports. Our proposed mechanism, designated clustering geo-indistinguishability, creates obfuscation clusters to aggregate nearby locations into a single obfuscated location. To evaluate the utility of the mechanism, we resorted to a real use-case based on geofencing. Our evaluation results have shown a suitable privacy-utility trade-off for the proposed clustering geo-indistinguishability mechanism.

Abstract
We propose a coding methodology for physical layer security with adaptive characteristics, whereby adaptive we mean that the system must be tunable to different operational points/signal-to-noise ratio levels of both the legitimate receiver and the eavesdropper. Based on interleaving and scrambling as techniques that shuffle the original message before transmission, we consider puncturing over an interleaving/scrambling key and/or over the message as a mechanism to provide the required adaptability to channel conditions. The proposed techniques have shown suitable adaptability to different channel quality levels of the legitimate receiver and eavesdropper, while still guaranteeing the desired reliability for the legitimate receiver and secrecy against the eavesdropper.

Abstract
We propose adding an irregular quadrature amplitude modulation (QAM) constellation to a wireless transmission scheme in order to obtain greater control over the signal-to-noise ratio (SNR) required to successfully decode the signal. By altering the separation between adjacent symbols, the minimum required SNR is raised without degradation in the performance of the scheme. This allows the system to adapt to preferable channel conditions for the authorized user, making it harder for eavesdroppers to intercept and decode the transmission, thus making the communication safer. In addition, we show that by overlaying a coset code onto the QAM constellation, a new, stronger security gap metric can be further improved. Results show the effectiveness of this strategy with an interleaved coding for secrecy with a hidden key (ICSHK) scheme.

Abstract
We propose an adaptive secrecy scheme using polar codes with random frozen bits for a general wiretap channel, in which to protect the data from a potential eavesdropper, part or all of the frozen bits are randomly generated per message. To assess the secrecy level of the proposed scheme, three types of decoding strategies are evaluated: a matching decoder which knows the positions of all inserted bits inside the blocklength and tries to estimate them using the same decoding techniques, a blind decoder which treats all the frozen bits as the same value, and a random decoder which considers those dynamic bits as random at the receiver. Results are presented in terms of the system security gap, assuming an adaptive decoding strategy. It is shown that the system achieves combined secrecy and reliability. The proposed scheme does not assume knowledge of the eavesdropper's channel when defining the indices of information and frozen bits.

Abstract
The burst of social networks and the possibility of being continuously connected has provided a fast way for information diffusion. More specifically, real-time posting allowed news and events to be reported quicker through social networks than traditional news media. However, the massive data that is daily available makes newsworthy information a needle in a haystack. Therefore, our goal is to build models that can detect journalistic relevance automatically in social networks. In order to do it, it is essential to establish a ground truth with a large number of entries that can provide a suitable basis for the learning algorithms due to the difficulty inherent to the ambiguity and wide scope associated with the concept of relevance. In this paper, we propose and compare two different methodologies to annotate posts regarding their relevance: automatic and human annotation. Preliminary results show that supervised models trained with the automatic annotation methodology tend to perform better than using human annotation in a test dataset labeled by experts.

Abstract
Sentiment lexicons are an essential component on most state-of-the-art sentiment analysis methods. However, the terms included are usually restricted to verbs and adjectives because they (1) usually have similar meanings among different domains and (2) are the main indicators of subjectivity in the text. This can lead to a problem in the classification of short informal texts since sometimes the absence of these types of parts of speech does not mean an absence of sentiment. Therefore, our hypothesis states that knowledge of terms regarding certain events and respective sentiment (public opinion) can improve the task of sentiment analysis. Consequently, to complement traditional sentiment dictionaries, we present a system for lexicon expansion that extracts the most relevant terms from news and assesses their positive or negative score through Twitter. Preliminary results on a labelled dataset show that our complementary lexicons increase the performance of three state-of-the-art sentiment systems, therefore proving the effectiveness of our approach.