Keywords:
Data mining.
;
Electronic books.
Description / Table of Contents:
Data mining is a rapidly growing research area in computer science and statistics. Volume 2 of this three-volume series covers theoretical aspects of the subject, including statistical, Bayesian, time-series and others relevant to health informatics.
Type of Medium:
Online Resource
Pages:
1 online resource (257 pages)
Edition:
1st ed.
ISBN:
9783642232411
Series Statement:
Intelligent Systems Reference Library ; v.24
URL:
https://ebookcentral.proquest.com/lib/geomar/detail.action?docID=885227
DDC:
006.3
Language:
English
Note:
Intro -- Title -- Preface -- Contents -- Advanced Modelling Paradigms in Data Mining -- Introduction -- Foundations -- Statistical Modelling -- Predictions Analysis -- Data Analysis -- Chains of Relationships -- Intelligent Paradigms -- Bayesian Analysis -- Support Vector Machines -- Learning -- Chapters Included in the Book -- Conclusion -- References -- Data Mining with Multilayer Perceptrons and Support Vector Machines -- Introduction -- Supervised Learning -- Classical Regression -- Multilayer Perceptron -- Support Vector Machines -- Data Mining -- Business Understanding -- Data Understanding -- Data Preparation -- Modeling -- Evaluation -- Deployment -- Experiments -- Classification Example -- Regression Example -- Conclusions and Further Reading -- References -- Regulatory Networks under Ellipsoidal Uncertainty - Data Analysis and Prediction by Optimization Theory and Dynamical Systems -- Introduction -- Ellipsoidal Calculus -- Ellipsoidal Descriptions -- Affine Transformations -- Sums of Two Ellipsoids -- Sums of bold0mu mumu KKKKKK Ellipsoids -- Intersection of Ellipsoids -- Target-Environment Regulatory Systems under Ellipsoidal Uncertainty -- The Time-Discrete Model -- Algorithm -- The Regression Problem -- The Trace Criterion -- The Trace of the Square Criterion -- The Determinant Criterion -- The Diameter Criterion -- Optimization Methods -- Mixed Integer Regression Problem -- Conclusion -- References -- A Visual Environment for Designing and Running Data Mining Workflows in the Knowledge Grid -- Introduction -- The Knowledge Grid -- Workflow Components -- The DIS3GNO System -- Execution Management -- Use Cases and Performance -- Parameter Sweeping Workflow -- Ensemble Learning Workflow -- Related Work -- Conclusions -- References -- Formal Framework for the Study of Algorithmic Properties of Objective Interestingness Measures.
,
Introduction -- Scientific Landscape -- Database -- Association Rules -- Interestingness Measures -- A Framework for the Study of Measures -- Adapted Functions of Measure -- Expression of a Set of Measures -- Application to Pruning Strategies -- All-Monotony -- Universal Existential Upward Closure -- Optimal Rule Discovery -- Properties Verified by the Measures -- References -- Nonnegative Matrix Factorization: Models, Algorithms and Applications -- Introduction -- Standard NMF and Variations -- Standard NMF -- Semi-NMF (semiconvex) -- Convex-NMF (semiconvex) -- Tri-NMF (triNMF) -- Kernel NMF (LD2006) -- Local Nonnegative Matrix Factorization, LNMF (sparse1,sparse3) -- Nonnegative Sparse Coding, NNSC (coding) -- Spares Nonnegative Matrix Factorization, SNMF (SNMF1,SNMF2,CNMF) -- Nonnegative Matrix Factorization with Sparseness Constraints, NMFSC (NMFSC) -- Nonsmooth Nonnegative Matrix Factorization, nsNMF (nsnmf) -- Sparse NMFs: SNMF/R, SNMF/L (SNMF) -- CUR Decomposition (CUR) -- Binary Matrix Factorization, BMF (BMF,BMF2) -- Divergence Functions and Algorithms for NMF -- Divergence Functions -- Algorithms for NMF -- Applications of NMF -- Image Processing -- Clustering -- Semi-supervised Clustering -- Bi-clustering (co-clustering) -- Financial Data Mining -- Relations with Other Relevant Models -- Relations between NMF and K-means -- Relations between NMF and PLSI -- Conclusions and Future Works -- References -- Visual Data Mining and Discovery with Binarized Vectors -- Introduction -- Method for Visualizing Data -- Visualization for Breast Cancer Diagnistics -- General Concept of Using MDF in Data Mining -- Scaling Algorithms -- Algorithm with Data-Based Chains -- Algorithm with Pixel Chains -- Binarization and Monotonization -- Monotonization -- Conclusion -- References.
,
A New Approach and Its Applications for Time Series Analysis and Prediction Based on Moving Average of nth-Order Difference -- Introduction -- Definitions Relevant to Time Series Prediction -- The Algorithm of Moving Average of nth-order Difference for Bounded Time Series Prediction -- Finding Suitable Index m and Order Level n for Increasing the Prediction Precision -- Prediction Results for Sunspot Number Time Series -- Prediction Results for Earthquake Time Series -- Prediction Results for Pseudo-Periodical Synthetic Time Series -- Prediction Results Comparison -- Conclusions -- Appendix -- References -- Exceptional Model Mining -- Introduction -- Exceptional Model Mining -- Model Classes -- Correlation Models -- Regression Model -- Classification Models -- Experiments -- Analysis of Housing Data -- Analysis of Gene Expression Data -- Conclusions and Future Research -- References -- Online ChiMerge Algorithm -- Introduction -- Numeric Attributes, Decision Trees, and Data Streams -- VFDT and Numeric Attributes -- Further Approaches -- ChiMerge Algorithm -- Online Version of ChiMerge -- Time Complexity of Online ChiMerge -- Alternative Approaches -- A Comparative Evaluation -- Conclusion -- References -- Mining Chains of Relations -- Introduction -- Related Work -- The General Framework -- Motivation -- Problem Definition -- Examples of Properties -- Extensions of the Model -- Algorithmic Tools -- A Characterization of Monotonicity -- Integer Programming Formulations -- Case Studies -- Experiments -- Datasets -- Problems -- Conclusions -- References -- Author Index.
Permalink