« How to Apologize without Losing Your Pride | Home | Top tips on harnessing SEO power for your blog »
Do word relationship counselors obtain better search rankings? Google thinks so.
By Administration | September 28, 2007
Ah, the good ole’ days when a word was just a word. You need not wonder if you paid enough attention to it, brought it flowers on rare occasions, or remembered its birthday. Oh, how times have changed.
When searching Google, you may have recently noticed a considerable change in overall search results and relevancy scores. This has largely been contributed to what I refer to as Googles new virtual relationship counselor, but this is better known as Latent Semantic Indexing (LSI). What is LSI and how does it work, you ask? Well, the good news is that you need not make an appointment, pay $200 per hour, nor sit on a reclining couch and answer questions about your childhood to benefit from its services.
Latent Semantic Indexing essentially allows a search engine to ascertain the topic of a web page in a different way than simply matching search query text, or keywords. Rather, it calculates and places weight on related words within content. By performing the related words calculation, LSI will essentially lower the page ranking if it does not recognize related words to support the main keyword. There is a sensitive balance here, as to place too much emphasis on a keyword within an article can create over-optimization and this will have an adverse effect on overall ranking. LSI looks for the relationships between words, and if you have too many of one keyword without the proper word relationships, the site will be punished.
Let’s take a look at an example to better understand the importance and usage of Latent Semantic Indexing. A web page whose main topic is Fast Food would commonly and naturally have keywords such as french fries and hamburgers. LSI considers documents which have a great number of words in common to be semantically close. On the other hand, documents that have few words, if any, in common will be ranked as semantically distant, thus a lower ranking would be rendered. It should be noted that the LSI algorithm will take into consideration words that are located in a similar position or words that are found within related documents when formulating the semantically close / semantically distant determination.
The LSI algorithm may not fully understand the meaning behind the words that are within a document. However, it fully comprehends the patterns and relationships that form within and this generates a sophisticated and uniquely robust step to the indexing process.
So go ahead, consider yourself the Dear Abby of prose and counsel your words to form relationships!
Share ThisComments are closed.