Latent semantic indexing (LSI) is an indexing and retrieval method that uses a mathematical technique called singular value decomposition (SVD) to identify patterns in the relationships between the terms and concepts contained in an unstructured collection of text.
This is a technique used by search engines to help determine the ‘meaning’ of a web page to help bring better SERPs (search engine results pages) to the website visitor to help them what they are searching for easier. In the past, this was an important topic in the SEO world but the search engines have gotten so good at this that all you have to do is write good, relevant topics for your content and this will take care of itself.