• Search Engine Journal, Monday, March 3, 2008 11:45 AM

The majority of user-facing video search engines rely on user tags to index, crawl and return relevant results. If a user (or group of users) has wrongly identified a video, then it will likely show up for the wrong query. Pluggd tries to circumvent that issue by analyzing the actual video content and making it searchable by keyword.

Seattle-based Pluggd converts the audio in each video to text and then indexes it via keyword and context. Users that run a Pluggd video search will find a heatmap-style display of videos that pertain to their query. "The color code on the map represents the extent to which the content spoken in video matches the search query of the user," says Arun Radhakrishnan.

But there are still kinks to work out with Pluggd's video search--namely the fact that it takes time to process the wealth of video content on the Web so thoroughly.

Read the whole story at Search Engine Journal »