“The intuition behind ‘latent semantic analysis’ (LSA) is to find the latent structure of ‘topics’ or ‘concepts’ in a text corpus, which captures the meaning of the text that is imagined to be obscured by “word choice” noise. The term ‘latent semantic analysis’ has been coined by Deerwester et al. who empirically showed that the co-occurrence structure of terms in text documents can be used to recover this latent topic structure, notably without any usage of background knowledge.” Gregor Heinrich ( 2008 )