Jargon Buster: Latent Semantic Indexing (LSI) |
|
|
| Written by Simon Bennison |
| Wednesday, 01 April 2009 08:46 |
|
LSI is a system of indexing that looks for semantically similar words in order to measure the relevance of a page. Where search engine optimisation previously focused on repetitions of keywords, LSI scans entire documents to look for common themes, synonyms and meanings. There is already evidence that Google is experimenting with ranking sites in this way. You can test this yourself by performing a search with a ~ symbol before your keyword. For example, a search for ~Scotland returns results with semantically similar keywords highlighted in bold. Scotland is at the top of the results, but you can also see that Glasgow and Edinburgh have been returned as highlighted words in the search. Try the same for the search ~apple and you’ll see that the third and fourth hit for the search are from the Microsoft website, highlighting the word ‘Windows’. According to the traditional wisdom of search engine optimisation (SEO), optimising for the keywords ‘Glasgow’ or ‘Windows’ should require repeats of those keyword throughout the page. If the search engines fully embrace this method of indexing, this will revolutionise the search marketing industry. What do you think?
Bookmark
Email This
Comments (0)
![]() Write comment
|