jump to navigation

Review: Exploiting Social Networks for Internet Search October 4, 2007

Posted by shahan in online social networks.

Exploiting Social Networks for Internet Search, A. Mislove, K. P. Gummadi, and P. Druschel. HotNets 2006.

Of the three papers to read this week, this was by far the most interesting. Not only is it pertinent to my field of information retrieval, but it is the only one to derive results by conducting a real-world experiment. This article, which discusses a more social reason for the success in their social network search method, is in pleasant contrast to the previous required reading from Mislove et al., Measurement and Analysis of Online Social Networks, in which they conduct a battery of statistical analyses.

The focus of this paper is in the use of cached results from a connected group of individuals during their search for information. The authors demonstrate a 9% increase in the effectiveness of search results and attribute this to 3 reasons: disambiguation, ranking, and serendipity.

The paper encourages a deeper look into how large a “cluster” should be to exploit such advances in search effectiveness. In the paper’s experiment, the groups were relatively close and it will be a challenge to be able to discern groups on a larger scale especially since, as was described by Watts, 2 or 3 dimensionally independent categories are most effective in determining social relatedness. Unfortunately, the question of privacy is a very important issue and will most likely be the biggest stumbling block of putting this system into practice. This alone is a major challenge: to determine what level of social relatedness will allow someone to access a network tie’s previous searches. One possible solution is for a specialized group to offer their previous searches on a paying basis, thus becoming similar to a Google Answers system on a larger scale, a cognizant expert system if you will.



1. Seifu T. - April 27, 2008

Great comment Shahan.

I read the paper (the summarized version that I found at http://www.mpi-sws.mpg.de/~amislove/publications/PeerSpective-HotNets.pdf) and I think the idea of improving search based on information gleaned from social relatedness is great.

There are things I don’t understand, though:
-In the paper, it is described how the HTTP proxy installed on the computers of the peers voluntarily participating in the experiment was used to index the URLs visited by these users. The queries submitted to the PeerSpective system were matched against these indexes distributed in these peers (if I am not mistaken). I don’t thus see how this is different from just a distributed (decentralized) search system. If this was done in a group consisting multiple social circles, was there any mechanism stated for discerning to what social circle an individual belonged to? In the absence of that, how is their system different from a decentralized Web search engine?
-One of the problems of conventional search engines pointed out by the authors was the lack of indexing on non-html pages. How would the use of social network-based searching do away with this problem? I didn’t see any data related to this point in the results of the experiment. Did I miss something?
-Again I don’t see how the idea of improving results on pages in the deep web or dark web… relate to social network based search than to the search merely being distributed.
-Finally, the authors raised an interesting point in noting that the ranking of a search result can be fine tuned based on an individual’s “context” of the term- the context being derived from what social network the individual belongs to. However, while I thought this would at the same time reduce the “serendipity” of the information in the query results as the information would be defined in a narrow context, the authors contended the reverse. What do you think?

I am eager to hear your comments on each one of the above.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: