Reducing Dimensionality Using NMF Based Cholesky Decomposition
Jasem M. Alostad · 2017
This paper aims to resolve the problem associated with increased data dimensionality in datasets using modified Non-integer Matrix Factorization (NMF). Further, the increased dimensionality arising due to non-orthogonally from NMF is resolved using Cholesky decomposition (cd-NMF). The cd-NMF is used to extract the feature vector from the dataset and the data vector is linearly mapped from upper triangular matrix obtained from the Cholesky decomposition. The experiment is validated in terms of accuracy and normalized mutual information metrics again three different text databases with varied patterns. Further, the results proves that the proposed technique fits well with larger instances in finding the documents as per the query, than NMF, NPNMF, MM-NMF, RNMF, GNMF, HNMF and cd-NMF.