In today´s traditional search approach, when a user types a keyword for searching, the search engine searches through its collection of document for the typed keyword. It checks for the exact typed keyword in all the documents and returns only those documents which contained it and ranks it on some ranking algorithm.

Words which are same in meaning but spelled out differently would be totally ignored in the Search results (SERPs.) Plus the results retrieved would have 50% sites which are totally irrelevant to the search query

There are two fundamental problems in natural language processing : synonymy and polysemy . In synonymy, different writers use different words to describe the same idea. Thus, a person issuing a query in a search engine may use a different word than appears in a document, and may not retrieve the document. In polysemy, the same word can have multiple meanings, so a searcher can get unwanted documents with the alternate meanings.

To address this important issue of presenting only relevant search results, Latent Semantic Indexing or LSI was proposed as a new improved method of retrieval system.

Latent semantic indexing helps search engines to find out what a web page is all about. It basically means to you that you shouldn’t focus on a single keyword when optimizing your web pages and when getting links.

In our LSI E-Book, you can know about the origin of LSI, how Google and Applied Semantics started as Search Engine and history of Adsense and Applied Semantics and much more.

• Google buys applied semantics
• Adsense and applied semantics
• What is LSI
• LSI and search engines
• How Google adopted LSI
• Proof of Semantic Readings