Skip to main content



PageRank’s Alternative Applications

In the first article, PageRank is demonstrated to be not only an algorithm for searching, but also a method to create general recommendation systems. Since PageRank is used to rank nodes in a graph based upon network measures, it is logical to wonder what other applications it has in the real world. Often times, when consumers visit online stores such as Amazon, they can see various products in a ‘recommended’ section. The products shown are often based upon basic factors like category and price, but PageRank allows online stores to utilize deeper and more advanced factors. In the article, Netflix is used as an example. PageRank is used to determine whether or not a movie is relevant for a user if other similar users liked it (a user is similar to another user if they like movies that are relevant to the first). Thus, by using PageRank, the recommended movie section is much more personalized and relevant.

Another recommendation system is PageRank’s application to ranking biomedical literature by importance, as described in the second article. Based on a study from the University of Birmingham, researchers have shown that PageRank has the potential to be applied to bibliometrics. In the past, raw citation counts have generally indicated articles of higher importance. However, with more and more citations being added online, this method is becoming less optimal as citation importance is not part of the consideration. By constructing a citation network, researchers can better determine the best articles with greater ‘scientific gravity’, and thus save time and effort sifting through literature. 

 

http://web.stanford.edu/class/msande233/handouts/lecture8.pdf

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4674919/

Comments

Leave a Reply

Blogging Calendar

October 2019
M T W T F S S
 123456
78910111213
14151617181920
21222324252627
28293031  

Archives