Exploring the evolution of social contexts with time can provide unique insights into human social dynamics. Several social contexts and relationships can be mined from unstructured text articles that describe social phenomena. In contrast to structured graphs of social networks, named entity recognition is a task that attempts to classify elements in unstructured textual items into predefined categories, such as organizations, people, locations, quantities, and temporal expressions. State of the art systems have approached the quality of human annotators on static documents for multiple languages. The problem of constructing and linking corresponding entities across topics and documents still exists. During a temporal sequence, entities fluctuate in frequency over time, and the set of entities in the present document can differ from the beginning and end. Furthermore, with user-generated content available on most major news sites, different viewpoints and entity relationships are generated by different users. This paper describes the Sequencer system for the temporal analysis of named entities in news articles between media reported stories and user generated content.