Analysis of online video search and sharing

Abstract

It is now feasible to view video at home as easily as text-based pages were viewed when the Web first appeared. This development has led to the emergence of video search engines providing hosting, indexing and access to large, online video repositories. A key question in this new context is whether users search for media in the same way that they search for text. This paper presents a first step towards answering this question by providing novel analyses of people's linking and search behavior using a leading video search engine. Initial results show that page views in the video context deviate from the typical power-law relationships seen on the Web. However, more positively, there are clear indications that tagging and textual descriptions play a key role in making some video-pages more popular than others. This shows that many techniques based on text analysis could apply in the video context.