Using social networking and collections to enable video semantics acquisition

Video content both professionally and user generated is now a prime driver of Internet usage and activity; Simultaneously, the Internet has seen the rise of social networking sites and activity. This paper brings together video content and social networking to solve the problem of gathering metadata which describes user interaction, usage and opinion of video content. The paper explores mechanisms by which implied and explicit semantics can be garnered from users interacting with media within social networks. This leads to the gathering of individual user interaction metadata which is then aggregated to form semantic metadata for a given video. The techniques have been implemented in a custom Flex application which is based around the Facebook API. This provides various mechanisms for the collection and then display of descriptions of user interaction with video content.