LSI (Latent Semantic Indexing) isn’t exactly “new” or “bleeding edge” any more but it’s relevance and usage in optimizing one’s website should be and is a critical component of a sound SEO strategy. Said differently, if your website and search engine optimization project does NOT include an LSI strategy using the right software tools, your site will not be thematically sound from end to end. While it’s possible to get a little lucky when it comes to your content, I don’t like (and I know my clients don’t like) betting on luck for our results.
If you want thematically sound, and thus structured, content then LSI is the framework to achieve that. The search engines crawl your site simply for analysis only but LSI is part of the Google search algorithm because at the end of the day, having properly themed content, is exactly what users are looking for when they conduct a search. They are looking for information on a particular keyword phrase but ultimately they want a solid picture of the overall theme of what they are really after.
You’ve done it before too… you conducted a search on a specific set of keywords strung together. But in your search analysis, you discovered an even more specific keyword that got closer to what you were really after and thus you modified your search term based on what you found in the first set of results.
Google wants to return thematically structured content because one keyword has dozens if not hundreds of very closely related words and those groups of words (a theme) constantly are found together, in content as part of how we naturally write and relate words to other words.
If you can wind your words together around a close-knit theme within your content on your site, you will be miles ahead of the idiot who hires someone with broken English to write a 500 word article for his site.
After all, content (still) is KING. Quality, relevant content is what we are after folks. LSI is a big piece to solving that riddle.
More research you can do on your own below!
The same steps are used to locate the vectors representing the text of queries and new documents within the document space of an existing LSI index.
First, convert each document in your index into a vector of word occurrences. Now you have a Latent Semantic Index. The number of dimensions your vector exists in is equal to the number of unique words in the entire document set. Most document vectors will have large empty patches, some will be quite full.
Feb 4, 2005 • 8:57 am (2) by Barry Schwartz Filed Under Search Technology. Yesterday I wrote an entry named Latent Semantic Analysis (LSA). Crawl into the Google Algorithm. Where I discussed how the current theories behind the Google SERP changes have to do with a new algorithm shift for Google.
(LSI) in simple terms and without the college degree math that is usually required. In a follow up post I will explain why LSI is not used by search engines. April 7, 2007 at 2:38 pm. Filed under News and Comment.