Media Cloud, a new project from the Berkman Center at Harvard University, has an ambitious goal: It will do the heavy lifting of analyzing stories from thousands of traditional news sources, analyzing the semantics of the content through Calais (covered here and here), and then providing tools to quickly get trending results. This approach promises to bring what used to be an expensive and laborious process to anyone who has a need for this type of data but lacks the means to get it.
At launch, Media Cloud will offer three primary trend visualization tools and a discussion forum for anyone to use. In addition, a news RSS feed and mailing list are available. We'll now take a moment to review how these visualization tools work, and we'd also like to point you to a very illustrative video interview of project developer Ethan Zuckerman, by the Nieman Journalism Lab.
Disclosure: Reuters Calais is an RWW sponsor. And they are awesome.
Top 10 Chart
This tool lets you compare up to three media sources, generating a list of the top ten most mentioned terms for that source and relative frequency of use for each term. This chart can be useful in a number of ways, indicating not only what terms are considered most important by each news source at the moment you generate the chart, but also showing if there is a clear standout term that may indicate a very hot topic. Also, when comparing two similar media sources, say for example the New York Times and the Washington Post, the resulting chart can give you an idea of what each paper considers more important leading topics.
Top 10 Term Pivot Chart
You can put in your own search term and up to three media sources in this tool to see what terms are most frequently mentioned alongside the search term in those sources' stories. This allows you to gain insight on how frequently related terms cluster together. So, for example if you search for Obama, you might find that, while United States is the most common related term, CNN's focus is more on Congress while FOX News writes more about the White House.
World Map Chart
This tool shows global coverage of all terms in the Media Cloud database for the selected media sources. Naturally, a newspaper that is focused on national US news will not have the depth of coverage of a source that has an international perspective. But even when comparing similar international sources, the weight each source gives to news from different regions can differ greatly. Take the New York Times versus BBC news coverage, you will see that darker colors mean that BBC has heavier coverage of European affairs, while NYT has stronger coverage of Canada and Mexico news topics.
Media Cloud is a project that developed from discussions around where story trends came from. This tool attempts to serve as a foundation to help move these conversations forward, and the Berkman Center is keeping the door open for new ideas and ways of using this data. To that end, they also have a discussion forum where people can contribute suggestions, thoughts and ideas to the project. Media Cloud also provides an RSS feed and an email list you can subscribe to if you want to stay in the loop for any new developments.
Like our coverage of the New York Times R&D Labs, we see this as an example of how the Internet is driving traditional media to change and respond in new ways. We are excited by the scope and potential that Media Cloud brings to anyone interested in following news and media trends.