|AM Posters 2009||START Conference Manager|
More than half of online video viewers share links to the videos they find with others, and three in four say they receive links to watch video that others have sent to them. These online video viewers actively exploit the participatory features of online video, such as rating content, posting feedback and uploading videos. But what determines which videos will be shared, and by whom?
Why do some videos on the Internet 'go viral' while others languish in obscurity? What attributes of message format, message content and social network sharing contribute to the diffusion of video content through the internet? What motivates people to share video content with others? What type of content is most often shared, under what circumstances, and to whom? Is there a formula or pattern for what makes some media more viral?
This paper presents preliminary results from an Ontario Centres of Excellence funded project that is exploring media distribution through social networks. The overarching goal of the project is to identify the attributes of content and social network paths that are shared by the most popular and frequently shared videos on YouTube. This is considered a first step in developing a formula for determining the ‘virality’ of web videos overall.
The research builds upon earlier studies by Adami. (2007, 2008, 2009); Szabo. & Huberman (2008); Acharya, Smith & Parnes (2000); Cha, Kwak, Rodriguez, Ahn & Moon (2007).
The purpose of this phase of the study was twofold:
I. To identify facets that distinguish noteworthy videos on You Tube. Using the YouTube API, metadata associated with each of the top 25 videos from the categories: Most Popular, Most Viewed, Most Discussed and Most Responded.
Metadata for the top 25 videos in each category was collected for one week and a comparative .content analysis was conducted.
II. The second goal of this phase of the study was to map a 'niche network" of referring websites around a basket of popular Youtube videos within a narrow genre in order to find out what sites matter most to the niche network in terms of video referral behavior. If you could identify these sites and their network configuration, we argued, then you could start to measure when a video is being taken up by a niche network, a factor that would indicate virality.
In this paper we present preliminary results of our study of viral videos mapped within referral networks derived from YouTube ("sites linking to this video") and from Google (search return links for a unique video ID).
|START Conference Manager (V2.54.6)|