عنوان

Understanding complex datasets

9781584888321 (alk. paper)

IR

E-9507

انگلیسی

IR

Understanding complex datasets

[Book]

:data mining with matrix decompositions

/ David Skillicorn

Boca Raton

: Chapman & Hall/CRC Press,

, c2007.

xxi, 236 p., [8] p. of plates

: ill. (some col.)

; 25 cm.

(Chapman & Hall/CRC data mining and knowledge discovery series.)

Print - Electronic

Includes bibliographical references (p. 223-232) and index.

1Data Mining1 --1.1What is data like?4 --1.2Data-mining techniques5 --1.2.1Prediction6 --1.2.2Clustering11 --1.2.3Finding outliers16 --1.2.4Finding local patterns16 --1.3Why use matrix decompositions?17 --1.3.1Data that comes from multiple processes18 --1.3.2Data that has multiple causes19 --1.3.3What are matrix decompositions used for?20 --2Matrix decompositions23 --2.2Interpreting decompositions28 --2.2.1Factor interpretation -- hidden sources29 --2.2.2Geometric interpretation -- hidden clusters29 --2.2.3Component interpretation -- underlying processes32 --2.2.4Graph interpretation -- hidden connections32 --2.3Applying decompositions36 --2.3.1Selecting factors, dimensions, components, or waystations36 --2.3.2Similarity and clustering41 --2.3.3Finding local relationships42 --2.3.4Sparse representations43 --2.3.5Oversampling44 --2.4Algorithm issues45 --2.4.1Algorithms and complexity45 --2.4.2Data preparation issues45 --2.4.3Updating a decomposition46 --3Singular Value Decomposition (SVD)49 --3.2Interpreting an SVD54 --3.2.1Factor interpretation54 --3.2.2Geometric interpretation56 --3.2.3Component interpretation60 --3.2.4Graph interpretation61 --3.3Applying SVD62 --3.3.1Selecting factors, dimensions, components, and waystations62 --3.3.2Similarity and clustering70 --3.3.3Finding local relationships73 --3.3.4Sampling and sparsifying by removing values76 --3.3.5Using domain knowledge or priors77 --3.4Algorithm issues77 --3.4.1Algorithms and complexity77 --3.4.2Updating an SVD78 --3.5Applications of SVD78 --3.5.1The workhorse of noise removal78 --3.5.2Information retrieval -- Latent Semantic Indexing (LSI)78 --3.5.3Ranking objects and attributes by interestingness81 --3.5.4Collaborative filtering81 --3.5.5Winnowing microarray data86 --3.6Extensions87 --3.6.1PDDP87 --3.6.2The CUR decomposition87 --4Graph Analysis91 --4.1Graphs versus datasets91 --4.2Adjacency matrix95 --4.3Eigenvalues and eigenvectors96 --4.4Connections to SVD97 --4.5Google's PageRank98 --4.6Overview of the embedding process101 --4.7Datasets versus graphs102 --4.7.1Mapping Euclidean space to an affinity matrix103 --4.7.2Mapping an affinity matrix to a representation matrix104 --4.8Eigendecompositions110 --4.9Clustering111 --4.10Edge prediction114 --4.11Graph substructures115 --4.12The ATHENS system for novel-knowledge discovery118 --4.13Bipartite graphs121 --5SemiDiscrete Decomposition (SDD)123 --5.2Interpreting an SDD132 --5.2.1Factor interpretation133 --5.2.2Geometric interpretation133 --5.2.3Component interpretation134 --5.2.4Graph interpretation134 --5.3Applying an SDD134 --5.3.1Truncation134 --5.3.2Similarity and clustering135 --5.4Algorithm issues138 --5.5Extensions139 --5.5.1Binary nonorthogonal matrix decomposition139 --6Using SVD and SDD together141 --6.1SVD then SDD142 --6.1.1Applying SDD to A[subscript k]143 --6.1.2Applying SDD to the truncated correlation matrices143 --6.2Applications of SVD and SDD together144 --6.2.1Classifying galaxies144 --6.2.2Mineral exploration145 --6.2.3Protein conformation151 --7Independent Component Analysis (ICA)155 --7.2Interpreting an ICA159 --7.2.1Factor interpretation159 --7.2.2Geometric interpretation159 --7.2.3Component interpretation106 --7.2.4Graph interpretation160 --7.3Applying an ICA160 --7.3.1Selecting dimensions160 --7.3.2Similarity and clustering161 --7.4Algorithm issues161 --7.5Applications of ICA163 --7.5.1Determining suspicious messages163 --7.5.2Removing spatial artifacts from microarrays166 --7.5.3Finding al Qaeda groups169 --8Non-Negative Matrix Factorization (NNMF)173 --8.2Interpreting an NNMF177 --8.2.1Factor interpretation177 --8.2.2Geometric interpretation177 --8.2.3Component interpretation178 --8.2.4Graph interpretation178 --8.3Applying an NNMF178 --8.3.1Selecting factors178 --8.3.2Denoising179 --8.3.3Similarity and clustering180 --8.4Algorithm issues180 --8.4.1Algorithms and complexity180 --8.4.2Updating180 --8.5Applications of NNMF181 --8.5.1Topic detection181 --8.5.2Microarray analysis181 --8.5.3Mineral exploration revisited182 --9Tensors189 --9.1The Tucker3 tensor decomposition190 --9.2The CP decomposition193 --9.3Applications of tensor decompositions194 --9.3.1Citation data194 --9.3.2Words, documents, and links195 --9.3.3Users, keywords, and time in chat rooms195 --9.4Algorithmic issues196 --Appendix AMatlab scripts203.

DSU Title III 2007-2012.

Chapman & Hall/CRC data mining and knowledge discovery series

Data mining.

Data structures (Computer science)

Computer algorithms.

Matrices.

Decomposition (Mathematics)

Skillicorn, David B

ایران

Understanding complex datasets

عادی

عادی

E-9507.pdf

0

ایمانی

متن

0

E-9507

انگلیسی

old catalog

pe

BL

1

a

Y