Social networks are becoming one of the most popular forms of communication between individuals worldwide. As more people start to use social networks and post more status updates, more information about the personal lives of individuals begins to leak out into cyberspace. In this thesis, we leverage the power of social networks and their Application Programming Interface (API) to data mine social networks.

Many social networks are beginning to add geo location information to status updates to show where users post updates from. Using the geo location information we gathered from Twitter and Foursquare, we are able to analyze the spatial and temporal patterns of individuals. Using studied heuristics, we are able to predict the activity of these individuals. These predictions are perfect for recommendation engines, law enforcement applications, and to track the spread of disease and exposure of harmful pollutants in the atmosphere.

We use the spatial and temporal information from the social networks in combination with UCLA's Vehicular Sensor Network. We seek to answer the questions of where people go and how long do they stay in one location. The Vehicular Test-Bed currently deployed in Macao, China tracks the concentration levels of carbon dioxide in the Macao region continuously. Using the social network information, we are able to estimate the exposure of individuals to these harmful pollutants.