Trend mining in social networks: from trend identification to visualization

A four‐stage social network trend mining framework, the Identification, Grouping, Clustering and Visualization framework, is described. The framework extracts trends from social network data and then applies a sequence of techniques (‘tools’) to this data to facilitate interpretation of the identified trends. Of particular note is the visualization of trend migrations (changes) that feature within time‐stamped network data. The framework is illustrated using a sequence of four social networks extracted from the Cattle Tracing System in operation in Great Britain, although it could equally well be applied to other forms of temporal data. The presented analysis of the Identification, Grouping, Clustering and Visualization framework indicates advantages, with respect to network trend mining, that can be gained, especially when the framework is applied to real data.