Get semantically related keywords
Posted: Sun Feb 02, 2025 5:11 am
To understand this comment in a little more detail, let's take a look at this article by Bill Slawski on whether or not Google uses LSI.
Bill's analysis on the subject allows us to understand that, as we said before, LSI is an old technology (which actually existed before the web), and that if we often talk about it in the context of SEO, it is to explain how Google can understand synonyms and semantically related words.
Indeed, if we take a look at the original LSI patent , we can see:
Since human word usage is characterized by high synonymy saudi arabia telegram number database and polysemy, simple term matching systems have major flaws: relevant points will be overlooked because different people describe the same topic using different words, and because the same word can have different meanings, irrelevant points will be emphasized.
Simple term matching is not an effective way to index web content.
But just because Google is looking to understand semantically related words when indexing a web page doesn't mean that Latent Semantic Indexing is the technology being used. Quite the contrary, and among other evidence that Bill shares, there are Google texts on semantic topic modeling that don't reference LSI.
Yet, just because there is no evidence that Google is using Latent Semantic Indexing as a technology to understand the relationship between words and determine the topic of a page, doesn’t mean it isn’t being done in other ways. We can be confident that content that ranks high in SERPs usually covers a topic in sufficient depth and demonstrates authority and expertise.
Whether the term "LSI keywords" is correct or not, it's clear that you need to carefully consider the words you use to help Google understand what your content is about.
Bill's analysis on the subject allows us to understand that, as we said before, LSI is an old technology (which actually existed before the web), and that if we often talk about it in the context of SEO, it is to explain how Google can understand synonyms and semantically related words.
Indeed, if we take a look at the original LSI patent , we can see:
Since human word usage is characterized by high synonymy saudi arabia telegram number database and polysemy, simple term matching systems have major flaws: relevant points will be overlooked because different people describe the same topic using different words, and because the same word can have different meanings, irrelevant points will be emphasized.
Simple term matching is not an effective way to index web content.
But just because Google is looking to understand semantically related words when indexing a web page doesn't mean that Latent Semantic Indexing is the technology being used. Quite the contrary, and among other evidence that Bill shares, there are Google texts on semantic topic modeling that don't reference LSI.
Yet, just because there is no evidence that Google is using Latent Semantic Indexing as a technology to understand the relationship between words and determine the topic of a page, doesn’t mean it isn’t being done in other ways. We can be confident that content that ranks high in SERPs usually covers a topic in sufficient depth and demonstrates authority and expertise.
Whether the term "LSI keywords" is correct or not, it's clear that you need to carefully consider the words you use to help Google understand what your content is about.