In a paper titled,  Lexical and Semantic Clustering by Web Links, Filippo Menczer of the Dept. of Computer Science at Indiana University, sets out to empiracally test two conjectures proposed by his colleagues in the field of search engineering for the web. Menczer's paper reaches some practical conclusions that should be applied by search optimizers.

First off, let's describe the two conjectures Menczer's paper covers:

Link-Content Conjecture
This makes the assumption that pages which link to each other are closely topically related, i.e. a page on bananas is likely to link to other pages about bananas (or at least fruit). 

Link-Cluster Conjecture
This assumes that pages which are clustered together, or in the same "web-community" are closely topically related, i.e. the page about bananas from above links to a page about fruit which links to a page about food - all three are closely topically related.

In a thread at SEW, I comment on the paper and draw out some conclusions that can be put to use by SEOs as they optimize their sites & pages for the web. The applications generally backup suggestions made in the past:

• It is as important (or more important)to get links from on-topic TLDs (top-level-domains) as it is to get links from on-topic pages.
• Links (both internal & external) will be better if they remain on the subject matter of the page.
• Getting links from the same pages/sites that link to your top competition is critically important as it suggests you are part of the "web-community" on that subject.