Microsoft has an interesting project that they are presenting at the International Conference on Weblogs and Social Media, which is occurring right now (Microsoft’s project will be presented this Wednesday). Blews is a project that hopes to show what the political blogosphere tells us about news. Its goal is to move a step beyond typical news-aggregation sites, which usually sort news by categories, by identifying which news stories are the most frequently discussed political topics (user ays26 recently discussed Google News, a great example of a typical news-aggregation site). The argument is that the most frequently discussed story should be the most important story in politics, and therefore the most interesting story to the user. Blews shows the user the number of liberal and conservative blogs that link to a particular story, in addition to the level of emotional charge of the blogs. Another benefit of Blews is that it “enabl[es] a reader to compare different views on the same story from different sides of the political spectrum.”

In class, we discussed the link analysis ideas of hub-authority and PageRank that are used in the ranking functions of search engines. Blews appears to use similar link analysis methods to help determine the most important political story. In this way, Blews functions more like a standard search engine than typical news-aggregation sites, which rely on sorting news stories into topics. While the basic link analysis may be similar to a search engine, Blews limits itself to a very specific set of pages that link to news stories: political blogs. Also, Blews uses the textual content of each blog to determine its emotional charge, which I assume it uses to help weight the importance of that blog’s links. In other words, the higher the emotional charge of the blog, the greater likelihood that the story it links to is interesting. The news story with the most highly emotionally charged links is the most popular.

The idea of Blews is a novel one, in that it uses a select group of people to determine the most interesting story in politics. It appears to be a proof of concept, but I’m sure that more link analysis ideas will be incorporated into news-aggregation websites in the future.