Talk:Keyword density/Archives/2016

Untitled
google not using KD section Hi, I have add some content along with references to show that google does not and has not used keyword density in its ranking algorithms for some time. I have worked in seo for many years now and know this to be a fact, so if anyone wants to contest my additional and up to date edits to this page please feel free to message me and I will find you more sources of google saying they dont use keyword density and haven't done since about 2009! GSL-Nathan (talk) 11:49, 23 October 2012 (UTC)

Search Engines Do Not Use Keyword Density

Instead they use things like term frequency, on a local scale with in a document, and on a global scale within the corpus of documents within their databases. Bill Slawski 02:19, 1 June 2006 (UTC)

I agree they also use that. But "title", "h1" tags also matter. And for all practical purposes, it is also wise to have body text match above somewhat. Perhaps the topic should be weighted keyword density :) --Tom


 * "'Term frequency' on a local scale within a document' is keyword density. Search engines compare how many words are in a document, and how frequent one specific term appears in that document. What you described is exactly what keyword density is. JerseyGirlMedia (talk) 01:02, 15 July 2009 (UTC)


 * No, that isn't keyword density. See: tf–idf -  Bill Slawski (talk) 04:25, 20 July 2009 (UTC)


 * Actually, yes it is. The source you cited states that very clearly. Here is a quote from tf–idf: "Consider a document containing 100 words wherein the word cow appears 3 times. Following the previously defined formulas, the term frequency (TF) for cow is then (3 / 100) = 0.03." That's the same formula for keyword density. When you start considering the whole corpus of documents, then you get into inverse document frequency, and things change. Although search engines may not use keyword density alone, we have no way of determining what IDF value they may arrive at without access to their databases. So keyword density remains our best way of measuring the value of a particular keyword in a given document. Daeruin (talk) 19:22, 30 November 2010 (UTC)