In social networking websites people and connections are map ped out. These maps represent relationships. The names designated as “nodes” are in clusters with edges representing the many connections or relations a node has with another. Similarly with the video content the names serve as “nodes” which are classified into a specific cluster based on its relevance. The main idea of performing TDA at an early stage is to reduce large raw data sets into useful small areas without losing any important information or without losing any of the topological properties. Using mathematical number for classification enables us to analyze how similar or dissimilar nodes or groups are from each other which can then be used and encoded as a distance function. These compressed representations will eventually then declare the underlying shape of data. The video content can be classified without performing any hard computation.