Keyword Density, SEOs, and the Deception War
I’m happy that at this Sphinnessed post: http://www.searchenginepeople.com/blog/how-search-really-works-the-keyword-density-myth.html , several SEOs are finally waking up and getting the Keyword Density Myth.
Great to see that they are realizing what IR grad students already know (http://irthoughts.wordpress.com/2008/01/25/the-power-of-document-linearization/ ):
That KD is not even a cost-effective tool for detecting spam, as search engines can use local term weight models that in addition to scoring terms can attenuate word frequencies. More on this here http://irthoughts.wordpress.com/2007/05/09/keyword-density-the-devils-advocate/ and here http://irthoughts.wordpress.com/2007/05/07/keyword-density-kd-revisiting-an-seo-myth/
Such models effectively minimize spurious effects/advantages derived from keyword repetition. Some of these are LOGA, LOGN, ATF1, and SQRT. One could even use the idea of global ENTROPY and propose a local ENTROPY model to neutralize any attempt to misrepeating terms. All these have been discussed in my Web Mining course.
Based on the aforementioned links, it is clear that the Search Engine War never ends, especially when in addition to their spam tactics, marketers are proposing soooo many theories made out of thin air. What is worse is that from time to time they induce their peers and cheerleaders to buy into their Latest SEO Incoherences (”LSI”).
The recent round of nonsense surprisingly comes from alleged “SEO experts”. From claims about sculping PageRank (http://sphinn.com/story/26410 ) to the usual LSI non-sense (http://irthoughts.wordpress.com/2007/07/09/a-call-to-seos-claiming-to-sell-lsi/ ) to Keyword Density (http://irthoughts.wordpress.com/2007/12/20/from-keyword-density-to-william-tuttes-legacy/ ), the Deception War never ends.
I’m glad at AIRWEB we fight these folks (http://airweb.cse.lehigh.edu/2008/cfp.html ). Irronically, what Google’s Cutts calls “proper user” of nofollow link attributes and “sculping” is just another way of saying end user’s human manipulation attempts. Great pr exercises.
