More about LSI and SEO
Marie-Claire Jenkins (a.k.a CJ), the Author of Science for SEO has an excellent article How does a search engine know what words mean?
My favorite section from CJ’s post is:
“Language is very ambiguous (meaning can be misleading – more than one meaning) and so are the senses used by different people. It is not always possible for the machine to give a very precise solution because of this variance. In fact humans have been shown to be about 90% accurate. ”
It fascinates me as an SEO how contextual words can completely change the meaning of a single word. How human perspective can completely change a definition for a word.
CJ was also interviewed by Joe Hall at Marketing Pilgrim Where she explains Natural Language Processing in-depth for the layman. I didn’t realize search engines were using NLP since the 1950’s or how integrated it is in our everyday lives. Read the interview Understanding Natural Language Processing for SEO
What words are you missing?
Latent Semantic Indexing (LSI) and Natural Language Processing (NLP )is a very large part of how search engines work today. Wait! Don’t let your eyes glaze over just yet I am not going to engage in tech babble. My own explanation of LSI / NLPand its use in search engines:
As our language changes and slangs are added into the mix it is difficult to determine the relevancy and meaning of a word without the context it is used in. As an example take the reference ”911″ before 2001 we related to word to emergency phone numbers- police and fire fighters. Now it has a whole different meaning in our common language. So how is “911″ to be indexed? What search results should appear? This is where LSI and NLP technology shines.
Remember the back in the day when the name of the SEO game was how many times and ways you could insert a word. Now it is called keyword stuffing and can get you hit with the ban stick! And if you don’t remember those days then pay attention young grasshopper.
It takes a combination of words to equal relevance to a single term or phrase. But what if you don’t know which words on your page you are missing to be relevant for the term you are aiming for? You could spend hours sifting and comparing the words on competitor pages to yours, combing Dictionary.com for the synonymous words and phrases or you could have SEM Scout.
Michael Marshall has created SEM Scout, a desktop application that will show you what words you are missing using a very powerful (well tested) method among the many approaches of NLP. I have had so many AH HA moments using Scout I cannot imagine not having it at my fingertips for my next set of articles or SEO challenge. SEMScout is a tool that utilizes a unique approach to help SEO’s focus our pages to ensure the right context is in place to define our pages.
Our language constantly evolves even more so with the social media explosion and blending of cultures encourages by the ability the internet gives us to communicate across the globe instantly. We add new words to our dictionary every year. In 2008 the words added were: W00t, fanboy, webinar, netroots, and pretexting. What words do you think will be added in 2009?
