“I believe that the search engines analyze semantics so we don’t have to!”

…would be that because search engines analyse semantics (and beyond) then it’s our jobs to do so! I don’t mean to sound like an algorithm-chaser, but I think that SEO’s could do far worse that thinking about the interpreted meaning of everything they do in the content of all other forms of information, on and off-site.

Regarding the point about subtlety – I think this is very important, which is why thinking about semantics, regarding meaning, can go far beyond the sometimes crude use of keywords and the like. I think that I would prefer to keep one eye on quality content and one eye on SEO…I think it’s doable anyway! Cheers Glenn, always insightful input!

For SEO copywriting: http://www.divinewrite.com/

@CJ …I’ve just emailed you in response to your post…I think what you said is very intelligent thinking, and having read your blog I am really impressed.

Although we approach things from slightly different backgrounds – you with the scientific background and we with the marketing background – I think we’ve arrived at similar conclusions. That’s certainly a positive in my eyes.

p.s. having looked at your blog, I just wanted to recommend it to others…certainly worth a read! http://scienceforseo.blogspot.com/

Thanks for the mind blowing comment CJ.

To do this there are very different techniques in research in the areas of , yes semantics, but this is just a part of it. Syntactics are very important (thus the choice of grammar used), and AI techniques need to be used also for machine learning and query expansion, as well as query understanding.

Many in the field are working on natural language generation and understanding because it does look like the obvious direction to go in. Why determine the meaning and quality of a document and provide an antiquated approach such as a list of documents, which ultimately gives you resources and not the answer to your question. In fact in my work I have observed many people using natural language for hard and ambiguous queries, which is not really unusual to tell you the truth.

Classification is another area which is deemed rather important also, and this does rely on meaning. Documents are analysed for topic and then classified in a number of different areas (indexed too), but then the difference is that they no longer work with simple keywords in any form but the overall semantic scoring of the actual documents. Why discover the meaning of a document and then index and store via keywords?

You are right from the SEO perspective I think, but I would suggest looking at the overall intention of your site and making that very clear by using appropriate structures rather than looking at the word as the atomic unit.

This is why I keep battering on about SEOs finding out more about IR techniques, looking beyond the pre-processing tasks that are things like LSI.

I did like your article though and found it extremely relevant to things going on at the moment.

I believe that the search engines analyze semantics so we don’t have to! If you develop a helpful, informative site that’s targeted to prospective customers, it’ll naturally be filled with all sorts of words they’re searching for. The search engines are definitely smart enough to figure out what you’re all about.

Of course, that’s not to say we shouldn’t focus on keywords. They keep us focussed. But I think, from an optimization point of view, most people would do better to invest time in quality content and social media optimization than in the subtleties of semantics.

Just my two cents’ worth!

Cheers
Glenn (Twitter: @divinewrite)