Paper Reading #18 - News Browsing

Comments:
Comment 1
Comment 2

References:
Title: Aspect-level News Browsing: Understanding News Events from Multiple Viewpoints
Authors: Souneil Park, SangJeong Lee, and Junhwa Song
Venue: IUI 2010, Feb. 7-10 2010

Summary:
In this paper, the authors describe a method of alleviating media bias in the news by providing varied versions of stories on the same subject. They call this method aspect-level news browsing.

Their system involves partitioning articles about a certain topic into different quadrants depending on the article's subject matter. They do this by analyzing articles in two different ways. One way they analyze is by dissecting the article and examining the first paragraph, where journalists tend to cluster main information. The second way they analyze it is to analyze articles near the beginning and ends of events, since journalists tend to report on more diverse parts of the issue further after the event. They then do an examination on how good their system is by checking its results to other algorithms.

Discussion:
I think that this system is a great idea. I personally know family members who will not talk to each other because they are so politically polarized from each other. Hopefully with a system like this they could learn to analyze the issues further.

In reality though, I think that a system like this won't help because most people won't spend the time reading multiple articles. Most people will just read one article and move on, which completely ruins the point of the program.

(Image courtesy of: Frogtown blog)