首页 | 本学科首页   官方微博 | 高级检索  
     

一种受限非负矩阵分解方法
引用本文:黄钢石,张亚非,陆建江,徐宝文. 一种受限非负矩阵分解方法[J]. 东南大学学报(自然科学版), 2004, 34(2): 189-193
作者姓名:黄钢石  张亚非  陆建江  徐宝文
作者单位:解放军理工大学通信工程学院,南京,210007;解放军理工大学通信工程学院,南京,210007;东南大学计算机科学与工程系,南京,210096;江苏省软件质量研究所,南京,210096;东南大学计算机科学与工程系,南京,210096;江苏省软件质量研究所,南京,210096
基金项目:国家自然科学基金,国家重点基础研究发展计划(973计划),国家自然科学基金
摘    要:提出一种获取潜在语义的受限非负矩阵分解方法.通过在非负矩阵分解方法的目标函数上增加3个约束条件来定义受限非负矩阵分解方法的目标函数,给出求解受限非负矩阵分解方法目标函数的迭代规则,并证明迭代规则的收敛性.与非负矩阵分解方法相比,受限非负矩阵分解方法能获取尽可能正交的潜在语义.实验表明,受限非负矩阵分解方法在信息检索上的精度优于非负矩阵分解方法.

关 键 词:非负矩阵分解  受限非负矩阵分解  潜在语义  信息检索
文章编号:1001-0505(2004)02-0189-05

Constrained factorization method for non-negative matrix
Huang Gangshi Zhang Yafei Lu Jianjiang ,, Xu Baowen . Constrained factorization method for non-negative matrix[J]. Journal of Southeast University(Natural Science Edition), 2004, 34(2): 189-193
Authors:Huang Gangshi Zhang Yafei Lu Jianjiang      Xu Baowen
Affiliation:Huang Gangshi 1 Zhang Yafei 1 Lu Jianjiang 1,2,3 Xu Baowen 2,3
Abstract:A novel method, constrained non-negative matrix factorization, is presented to capture the latent semantic relations. The objective function of constrained non-negative matrix factorization is defined by imposing three additional constraints, in addition to the non-negativity constraint in the standard non-negative matrix factorization. The update rules to solve the objective function with these constraints are presented, and its convergence is proved. In contrast to the standard non-negative matrix factorization, the constrained non-negative matrix factorization can capture the semantic relations as orthogonal as possible. The experiments indicate that the constrained non-negative matrix factorization has better precision than the standard non-negative matrix factorization in information retrieval.
Keywords:non-negative matrix factorization  constrained non-negative matrix factorization  latent semantic relations  information retrieval
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号