Bad Ratings : Beneficial?

I looked at an study based article called "Negative Ratings Play a Positive Role in Information Filtering" by Wei Zeng, Yu-Xiao Zhu, Linyuan Lü and Tao Zhou. In the article they talked about how ratings have helped in filtering information out that is not quality data, and in turn has helped lessen information overload.
I found the article by searching "information overload" in the TAMU library article database. I then clicked on the link that lead me to the full text online. Once arriving at the page when the article was, there was a yellow sticky note in the right corner that declared the article was a peer-reviewed/scholarly article. I also noticed that the article had an abstract and a reference section. These sections of the article confirmed that it was scholarly.
The authors support what they refer to as the Recommender System. This system is something we are all aware of even if we do not realize it. For example, when you go to Amazon and have previously looked at things on the site, the website will suggest other products you might like. This cuts down on the continuous search among all the products, which in the end causes information overload to be cut down also. They also say that the rating system helps with this problem too. When someone tells the server they do not like something, the server can then only dish out the information that is relevant to the internet surfer.
They come to the conclusion that "negative ratings are more significant for sparse data sets and for small-degree users/objects." This means that it is useful to those of us that use the internet for entertainment or just for every day minor activities. However, negative ratings can also lead to having future information hidden from you when you later need it or your preference changes. For normal users this may not be a big deal, but for those who use the internet for business or every day challenges this may cause an issue later down the road.
The article is not really trying to convince you of anything particularly. It is a study based article. They find their results and display the results to the audience in a very clear way. They do not just display the results that support their hypothesis they look at both sides. 
In the end, there is a lot of work to be done to help with the reduction of information overload. However, there are little steps being made currently to help with reaching this lofty goal.