News Distribution Network Raises Its Political I.Q. With Nexidia
Editor's Note: This is a vendor-written case study. StreamingMedia.com accepts vendor-written case studies based solely on editorial quality and their level of interest to readers.
In this age of real-time, breaking news coverage on the web, rapid access to video footage makes it easy for people to consume and share current events as they happen and for commentators to include source material as context to their discussions. News Distribution Network, Inc. (NDN) aggregates and distributes video from major news sources to internet sites. Leveraging Nexidia Video Enrich, a search tool that indexes videos by their audio content, NDN implemented an automated process to generate comprehensive metadata and enable keyword search and navigation, enabling the network to organize, syndicate, and monetize video from multiple news sources.
The Stakes Are High
NDN wanted to aggregate all content sources from a particular category and deliver them directly to the public via the websites people already visit. As the 2008 election season began to heat up, the network identified political content as an ideal starting point because it presented a vastly changed political environment driven by nontraditional media distribution. The political vertical has seen video employed as a viral tool combining content from not only politicians but also lobbyists, lawyers, special interest groups, and corporations. With a target market in excess of 160 million potential users, it’s not just about elections; issues, awareness of current events, and educating the public about the legislative process drive political video as well.
Acknowledging the need to aggregate this content, NDN developed Political I.Q. (www.politicaliq.com) not as a news portal but as an aggregator of video from several content providers. The site began to form relationships with major news organizations such as CBS, The Wall Street Journal, the Associated Press, AFP, Politico, and the National Journal as well as many elected officials, candidates, parties, and special interest groups.
Many political entities and content sources leverage YouTube or its competitors to share video content, but often, these videos are not tagged or sorted by the issues addressed within them, and serious content is often intermingled with irrelevant, satirical, or comical content, negatively impacting its credibility and usefulness. The speakers in Political I.Q. videos mention many topics—including the environment, energy, and Iraq—but usually, the video description is generic or there is sparse metadata, preventing these terms from showing up in a search. The challenge with the rapidly growing use of video content in the political process is figuring out how to deliver products that rapidly organize content and engage users. NDN needed the ability to let its visitors search within and organize videos by issue, candidate, political race, geographic location, and time frame, not just video title or a limited number of keywords.
With the goal of launching Political I.Q. prior to the 2008 presidential debates, NDN required a solution that could be implemented quickly, would mesh with its workflow, could be deployed on-site, and would work out of the box to analyze the content of more than 30,000 videos from dozens of sources to create comprehensive, standardized keywords and metadata. Rich information about the issues addressed in each video is essential when bloggers, news organizations, and other media outlets come to NDN to find content relevant to the current issues they are covering, which is then syndicated on their websites. For this, the network turned to Nexidia.
The Rich Media Search Solution
Nexidia Video Enrich delivers a foundation upon which content aggregators such as NDN may enhance and standardize the often disparate quality of metadata that accompanies rich media content.
Eschewing the labor- and time-intensive processes of editing clips, transcribing dialog, and manually defining metadata, Nexidia’s patented technology offers a rapid and automated process by which rich media content is rendered searchable, producing a phonetic index of the audio. Phonetic indexing provides a means of describing and classifying virtually all the sounds that can be produced by the human voice, and Video Enrich searches using phoneme pattern matching, which can be executed on blended words, phrases, proper names, slang, code words, brands, nonstandard grammar patterns, and ad hoc use of different languages. Deployed on a standard server, the indexing and tagging doesn’t take long—30 minutes of video can be indexed in less than 6 seconds.
Video Enrich also allows for the inherent flexibility of the human language. Phonetic searching emphasizes how things sound rather than their proper spelling, which is especially useful when searching for proper names, new words, entities, or phrases that are entering common speech but may be commonly misspelled.
This was precisely the dynamic engine NDN was seeking. Up and live within 2 weeks, Video Enrich lets NDN index video as it becomes available and deliver it within minutes of its release by the content owner.