LSI潜在语义信息检索模型

本文介绍了基于向量空间的信息检索方法 ,检索词和文件之间的关系表示成一个矩阵 ,查寻信息表示为检索词权重的向量 ,通过求查寻和文件向量的夹角余弦确定出数据库中的相关文件 .使用矩阵的 QR分解和奇异值分解 ( SVD)来处理数据库本身的不确定性 ,本文的目的是说明线性代数中的基本概念可以很好解决信息检索 ( IR)问题

LSI Latent Semantic Indexing Model

In the paper, a information retrieval method based on concept of vector space is introduced. A matrix represents the relationships between terms and documents, a query information is written as vector of weighting of terms, A document is returned as relevant according to the cosine of the angle between the query and document vectros in the database. QR factorization and singular value decomposition provide mechanisms for handling uncertainty in the database itself. The purpose of this paper is to show how fundamental mathematical concepts from linear algebra can be used to manage in information retrieval.