Evaluating the influence of events on mobility-based data communities

Human society can be divided into communities composed of individuals who share similarities in behavior, interest, housing, and other features. These human communities are affected by events, such as advertising, and traffic accidents, among others, daily. These events may affect communities differently, accordingly to each community's characteristics. This paper proposes an approach to assess the spatial-temporal influence these events exert on different geolocation-based social communities, thus contributing to planning actions focused on these communities. In addition, this paper contributes to a better understanding of people's and communities' behavior once their mobility patterns are analyzed. We carried out a case study using the Geolife dataset, a trajectory dataset, to accomplish the above contributions. Thus, using a clustering technique, dataset users were divided into communities based on their geolocation history. For each community identified, we evaluated the degree of influence of events in eleven places of good circulation on different days of the week.