The new SEO – Part 1: Latent Semantic Indexing

As you are probably aware, the search engines, especially Google, keep changing their algorithms in order to find the perfect match between visitor expectation and relevance of the pages within their index.

Like many acronyms, LSI and LSA have no meaning until you know what words those initials stand for. They sound like the names of popular television crime dramas, such as the fictitious “LSI: Miami” or “LSA:London.”

That’s not such a bad analogy, because LSA and LSI are mysteries to most Internet marketers. Not only that, but failure to use these important concepts would be a crime! Let’s try to get to the bottom of these unsolved mysteries.

LSI means Latent Semantic Indexing. LSA means Latent Semantic Analysis. That didn’t help much, did it?

Of course not. LSI and LSA are concepts that are better understood through examples. Before looking at examples, however, let’s look for clues in the words that make up these acronyms.

Latent Semantic Indexing

Definition of Latent:

Latent means “dormant, quiescent, not readily visible, or hidden.” It refers to qualities, abilities, or talents that may be currently hidden or unknown, but which might be revealed in the future.

Examples

1. His latent musical talent didn’t emerge until he had been to several piano lessons.

2. Her latent mathematical skills surprised her parents when she began bringing home very high test scores.

3. The car’s latent defects did not appear until the warrantee expired. All these examples refer to qualities or properties (both good and bad) that were always present, but took a while to become noticed.

Definition of Semantic:

Semantic refers to “the meaning, or an interpretation of the meaning, of a word, sentence, or document, and its relationship to other words in the text.” It does not mean the popular antivirus company (that’s Symantecâ„¢). It is also the study of meaning in language, or even the study of multiple meanings that a single word might have. The semantic aspect of a sentence, paragraph, or article is what it means to the reader.

Examples:

This word is most often used in its plural form “semantics,” as in “Let’s not argue over semantics.”

An example of arguing over semantics would be two people discussing the meaning of he word “war.” The first person insists war means the absence of peace. The second person is certain that war means the presence of combat. Both are correct in a sense, but it is like the argument whether the glass is half-full or
half-empty. Both parties are right about the meaning of war, but there is some interpretation involved.

For our purposes, “semantic” refers to the meaning of a word, sentence or document, how the reader interprets them, and the relationships between the words in the text.

Definition of Indexing:

An index is an “an alphabetized list of names, places, and subjects treated in a printed work, giving the page or pages on which each item is mentioned.” It is a tool for organizing printed material so that certain items in that material can be easily found again.

Examples:

The index of a book is an excellent example. If you read a book on internet marketing and wanted to review the information on affiliate programs, the book’s index would tell you on which pages the word “affiliate” appeared. You would quickly find the information you need.

Putting It All Together

Latent – refers to qualities or properties that were always present, but took a while to become noticed.

Semantic – refers to the meaning of a word, sentence or document, how the reader interprets them, and the relationships between the words in the text. Indexing – creating a list or reference for a printed work so that items within the text can be quickly found.

Therefore, Latent Semantic Indexing or LSI means a list or reference for printed material containing both the obvious and the hidden meanings and interpretations of the words, sentences, and paragraphs in that material, as well as how the words in the text are related to each other.

Latent Semantic Analysis

The definitions for latent, semantic, and indexing were presented above. All we need is a definition for “analysis.”

Definition:
Analysis – a detailed examination of elements or structure.

Examples

The best examples are the explanations you just read! The acronyms LSI and LSA were just examined. The words corresponding to each letter in the acronyms were determined, as well as the meanings of those words. The meanings of the individual words were then combined into a whole or index containing the
author’s interpretation of the concepts of LSI and LSA.

Therefore, Latent Semantic Analysis or LSA means an examination of the words, sentences, paragraphs, and the relationships between words in the printed material for obvious and hidden meanings. LSA must occur before the results can be organized and tabulated by LSI.

Why Is LSA and LSI So Important For You?

LSA and LSI are important concepts because search engines, especially Google, are beginning to analyze online documents using these techniques. Let’s restate that: Google and other search engines will be using LSA and LSI to read YOUR web pages.

Therefore you should know and understand what these concepts are. Once you do, you’ll be able to write web pages using LSI techniques. That will make the search engines happy with you and your websites. Hopefully, that will translate into more visitors and more profits.

In the offline world, LSA and LSI are techniques and concepts developed decades ago for analyzing printed documents. LSA is a technique used in natural language processing in order to analyze the relationships that exist between a specific set of documents, and the terms that are contained within those documents.

This is done by determining a main concept that relates to the terms within the documents, and that relates to the documents themselves. Once the LSA is complete, the results are compiled in an index by using LSI. The concepts and procedures are the same, whether online or offline.

Language is no barrier to LSA and LSI. If two sets of documents are written in different languages, they can be analyzed via LSI and catalogued by LSI if one set can be translated to the other set’s languages.

The task becomes easier of both sets of documents are written in the same language, which does not necessarily have to be English.

What does all this mean to you?

As a final consequence, it means for you that the days of keyword stuffing are over! If that’s what you’re currently doing to optimize your web pages, you should stop it now. Instead, make use of synonyms for the same words and try to find related words which you can use on your pages. This will make the theme “stick” and keep the search engines AND visitors happy!