Dear Forum Friends! I don't understant the Google Semantically Related Words & Latent Semantic Indexing Technology. Can anyone explain me how it work and how can we use it
LSI algo is implemented by most of the SE's, if you type "LSI" in DP you will find related thread where this topic is already discussed or google it.. Thanks!
Google doesn't use latent semantic indexing, neither does any other search engine. They don't have the capability. Latent semantic indexing is an algorithm to evaluate the relationship of every word to every other word. What Google does use is much cruder semantic matching. They taken out 8 or so patents over the years, and have mainly used it in adwords. If you want to judge how crude it is, look at what happens in a google adwords broad match. How do you use it? Well, the point of semantic matching in natural search is to evaluate the relationship of a keyword to synonyms and related words as they occur in normal written English (or whatever language is under consideration). So the answer is, write well and naturally.
I don't think that any of us here truly knows to what extent G is using LSI/LSA. It is a method that uses mathematical formulas for scoring documents and the terms within those documents. Related words are the key. So if you had a document about water and vacations, one could expect to find other terms like "beach, swimming, pool, hotel" and so on. If your document repeated "water" and "vacations" to the point where they are definitely the document's keywords, but you had no use of the other "expected" words, LSI would likely score your document low. I believe that G does in fact use LSI. This is very evident in the golden SEO rule "get links from related websites". LSI is all about relationships between terms. /*tom*/
Latent Semantic Indexing allows indexing search results based on the pairing of like words within documents. Many complex searches may lack exact matches in the results as well. Being able to find near matches will allow search engines to provide more comprehensive results.