摘要
潜在语义分析(Latent Semantic Analysis,简称LSA)通过奇异值分解(Singular Value Decomposition,简称SVD)分析文本集之间的关系,是产生关键词———语义之间映射规则的方法。而随后又出现的PLSA(ProbabilisticLatent Semantic Analysis)对基于奇异值分解的LSA又进行统计学的极大似然估计重新解释。LSA最初应用在文本信息检索领域,随着应用领域的不断拓展,LSA在信息过滤、跨语言检索、认知科学和数据挖掘中的信息理解、判断和预测等众多领域中得到了广泛的应用。
Latent Semantic Analysis provides a means of creating the mapping rule of key word - concept. And then the Probabilistic Latent Semantic Analysis （PLSA） gives the LSA wlfich is based on singular value decomposition a novel statistical explanation of maximum likely-hood, Initially, LSA is applied in the field of text information retrieval. With the ceaselessly development of its application, LSA has got a wide application field in Information Filtering, Cross-language retrieval., Cognitive Science and Date Mining which relate with information comprehension, judgment and prediction.
出处
《现代情报》
北大核心
2006年第11期205-206,共2页
Journal of Modern Information
关键词
潜在语义分析
PISA
奇异值分解
Latent Semantic Analysis
Probabilistic Latent Semantic Analysis
Singular Value Decomposition
