基于LSI的图书馆个性化信息服务系统的设计与研究

潜在语义索引（LSI）是近年发展起来的一种新的信息检索方法，本文以潜在语义索引技术为基础，从图书馆的个性化服务理念入手，介绍了图书馆个性化服务的现状，提出了图书馆个性化服务模型，详细讨论了应用潜在语义索引技术来建立用户个性化模型的系统结构和过程。实验中，我们选用KNN分类算法作为个性化信息的识别方法。通过分析比较表明，LSI是一种更有效的个性化特征选择方法；基于LSI的个性化文本和信息识别具有更高的精度。

Design and Research of a LSI-Based System for Individualized Information Services of Libraries

Latent Semantic Indexing (LSI) is a novel approach to information retrieval. This paper presents a model for individualized services in libraries based on LSI. It begins with the idea of individualized service, introduces its status quo,and discusses the system and the process to build an individualized model for the users based on LSI.In our experiments, we take the KNN algorithm as an identification approach, and the results show that LSI is a more effective method for individualized feature selection,and the LSI-based individualized text and information recognition has a higher precision.