Latent Semantic Indexing (LSI) and TREC-2

Latent Semantic Indexing (LSI) is an extension of the vector retrieval method (e.g., Salton & McGill, 1983) in which the dependencies between terms are explicitly taken into account in the representation and exploited in retrieval. This is done by simultaneously modeling all the interrelationships among terms and documents. We assume that there is some underlying or "latent" structure in the pattern of word usage across documents, and use statistical techniques to estimate this latent structure. A description of terms, documents and user queries based on the underlying, "latent semantic", structure (rather than surface level word choice) is used for representing and retrieving information. One advantage of the LSI representation is that a query can be very similar to a document even when they share no words.