Understanding and modeling the mobility of individuals is of paramount importance for public health. In particular, mobility characterization is key to predict the spatial and temporal diffusion of human-transmitted infections. However, the mobility behavior of a person can also reveal relevant information about her/his health conditions. In this paper, we study the impact of people mobility behaviors for predicting the future presence of flu-like and cold symptoms (i.e. fever, sore throat, cough, shortness of breath, headache, muscle pain, malaise, and cold). To this end, we use the mobility traces from mobile phones and the daily self-reported flu-like and cold symptoms of 29 individuals from February 20, 2013 to March 21, 2013. First of all, we demonstrate that daily symptoms of an individual can be predicted by using his/her mobility trace characteristics (e.g. total displacement, radius of gyration, number of unique visited places, etc.). Then, we present and validate models that are able to successfully predict the future presence of symptoms by analyzing the mobility patterns of our individuals. The proposed methodology could have a societal impact opening the way to customized mobile phone applications, which may detect and suggest to the user specific actions in order to prevent disease spreading and minimize the risk of contagion.

The wide adoption of mobile devices and social media platforms have dramatically increased the collection and sharing of personal information. More and more frequently, users are called to make decisions concerning the disclosure of their personal information. In this study, we investigate the factors affecting users’ choices toward the disclosure of their personal data, including not only their demographic and self-reported individual characteristics, but also their social interactions and their mobility patterns inferred from months of mobile phone data activity. We report the findings of a field study conducted with a community of 63 subjects provided with (i) a smart-phone and (ii) a Personal Data Store (PDS) enabling them to control the disclosure of their data. We monitor the sharing behavior of our participants through the PDS and evaluate the contribution of different factors affecting their disclosing choices of location and social interaction data. Our analysis shows that social interaction inferred by mobile phones is an important factor revealing willingness to share, regardless of the data type. In addition, we provide further insights on the individual traits relevant to the prediction of sharing behavior.

Mobile devices, sensors and social networks have dramatically increased the collection and sharing of personal and contextual information of individuals. Hence, users constantly make disclosure decisions on the basis of a difficult trade-off between using services and data protection. Understanding the factors linked to the disclosure behavior of personal information is a step forward to assist users in their decisions. In this paper, we model the disclosure of personal information and investigate their relationships not only with demographic and self-reported individual characteristics, but also with real behavior inferred from mobile phone usage. Preliminary results show that real behavior captured from mobile data relates with actual sharing behavior, providing the basis for future predictive models.

Online social networks like Facebook recommend new friends to users based on an explicit social network that users build by adding each other as friends. The majority of earlier work in link prediction infers new interactions between users by mainly focusing on a single network type. However, users also form several implicit social networks through their daily interactions like commenting on people's posts or rating similarly the same products. Prior work primarily exploited both explicit and implicit social networks to tackle the group/item recommendation problem that recommends to users groups to join or items to buy. In this paper, we show that auxiliary information from the useritem network fruitfully combines with the friendship network to enhance friend recommendations. We transform the well-known Katz algorithm to utilize a multi-modal network and provide friend recommendations. We experimentally show that the proposed method is more accurate in recommending friends when compared with two single source path-based algorithms using both synthetic and real data sets.