There is such a thing as InfluenceRank and it is relevant to social media as a whole. It is very difficult to rank users, find authority blogs and to create any sort of order in this world as a whole. It is important though because existing ranking algorithms are mostly made for digital libraries and search engines for example, but these don’t work well on blogs and other social media outlets.
I wote a couple of posts covering research in this area already:
Blogosphere vs Web – ranking issues
InfluenceRank:
I’m going to mention two papers which introduce the creation of InfluenceRank and the tests carried out on the method. Both papers are by Xiaodan Song, Yun Chi, Koji Hino, Belle L. Tseng from NEC laboratories who coined the phrase and invented the algorithm. Both papers are freely available to you as well.
This is the original paper where InfluenceRank was presented. It’s called “Identifying opinion leaders in the blogosphere“. I covered this already back in 2008, you can find the full summary on science for seo here.
I won’t go into the equations or the more in depth science in that paper because the whole point of this blog is to make these things as accessible as possible and as comfortable as possible for those who don’t need to give themselves a headache understanding it all. The high level concepts are the most important for online marketing. Those of you computer sciences can go ahead 
This is the abstract of that particular paper:
“Opinion leaders are those who bring in new information, ideas, and opinions, then disseminate them down to the masses, and thus influence the opinions and decisions of others by a fashion of word of mouth. Opinion leaders capture the most representative opinions in the social network, and consequently are important for understanding the massive and complex blogosphere. In this paper, we propose a novel algorithm called InfluenceRank to identify opinion leaders in the blogosphere. The InfluenceRank algorithm ranks blogs according to not only how important they are as compared to other blogs, but also how novel the information they can contribute to the network.”
The sequel:
The second paper is called “Summarization System by Identifying Influential Blogs“. It’s a short paper of only 2 pages and is written in an educated but very understandable way, so it will be accessible to all of you.
“InfluenceRank algorithm, in which blogs are ranked by how important they are to other blogs as well as the novelty of the information they contribute to the network. Information novelty of one blog is measured as the average information novelty of the entries in this blog compared to other entries which link to them”.
A query is issued to the system, then blogs that are deemed relevant to it are retrieved. Following this the blogs are ranked in order of how influential they are to other blogs and how innovative the opinions are in it. Then a summarization is created using those results.
SEO community take:
The SEO community has been talking about InfluenceRank as well and a nice post at SEP mentions it and brings up the HITS algorithm:
“Google’s pending patent basically applies their same principles of search to social. The first thing that comes to mind is the HITS algorithm which deciphers the hub and authority value of a domain. They’re essentially applying the same conceptual methodology to social media by applying the hub and authority value to a user instead.”
(A link to the pending patent would be welcome if anyone has it)
But it has been tested and it doesn’t work, as the paper discussed above demonstrates:
“As pointed out in ["Blogrank: ranking weblogs based on connectivity and similarity features"] , blog sites in the blogosphere are very sparsely linked and it is not suitable to rank blog sites using Web ranking algorithms like PageRank and HITS. The Random Surfer model of webpage ranking algorithms [43] does not work well for sparsely linked structures. The temporal aspect is most signi¯cant in the blog domain. While a webpage may acquire authority over time (its adjacency matrix gets denser), a blog post or a blogger’s inlfuence diminishes over time. Consequently, the adjacency matrix of blogs (considered as a graph) will get sparser as thousands of new sparsely-linked blog posts appear every day”.
David Harry has a good round-up of all the goings-on in social research on his blog. I would point out, as he does, that the Google patents are about content and context extraction, not about social ranking.
Better than PageRank:
“Compared to PageRank, our InfluenceRank algorithm selects more influential blogs with novel information. As an example, given query “YouTube”, http://www.captainsquartersblog.com/mt/ is ranked as 2nd by PageRank, while it gets demoted to 6th by our InfluenceRank algorithm because its information novelty is relatively low (0.73)”.
“Intuitively, top-ranked blogs detected by InfluenceRank are novel information contributors, and thus they tend to be scattered instead of focused in the network. As a result, important blogs detected by InfluenceRank tend to have a better coverage comparing to those detected by PageRank.”
And so now we know…
…that there is indeed such a thing as InfluenceRank, it is very real. The methods being described are valid, and tested. More testing and evaluation is always necessary but I have quite high hopes for this one or evolutions of it.
More research on ranking in social media:
“A few bad votes too many?: towards robust ranking in social media”
“Blog analysis and mining technologies to summarize the wisdom of crowds”
“Profile-Based Focused Crawler for Social Media-Sharing Websites”
(you’ll have to look for the free versions if any are available)
FriendRank:
Yes, it also exists and it was presented by researchers at Osaka University:
FriendRank: Friend Recommender System for SNS
A Basic Study on Friend Recommendation for SNS
This seems like this is a much better take on FriendRank than I noted in other posts but it is still plain wrong because nowhere in the patent that is linked to does Google even mention FriendRank. To top it all, that patent is for advertising.
In the Social Times post it says:
“So how is Google going to leverage their new “FriendRank” (a name already trademarked by social ad network, SocialMedia) patented technology?”
It’s already patented, and there is no evidence of it being used at Google. I may be wrong and that’s cool, but please point me to some resources. As far as I can tell any evidence of this type of thing is aimed at ads, not social media ranking.
Great post! thanks for the blog mention, glad to see follow up with this.
Pleasure Jordan, feel free to come write over here too should you get the itch!