Analysis and study on latent semantic index retrieval model
隐含语义索引模型的分析与研究

Latent Semantic Index (LSI) retrieval model was designed by expanding Vector Space Model. Based on the profound research of Vector Space Model, LSI mapped documents and terms vectors into a lower dimensional space by Singular Value Decomposition (SVD), so that the inherent vagueness associated with a retrieval process based on keyword sets was considerably reduced and semantic association among the documents was highlighted consequently. Theoretic analyses show that LSI can improve retrieval performance significantly.