1 / 17

Understanding Data Mining and Visualization: Techniques and Applications

This document explores the relationship between data mining and visualization, emphasizing how visualization aids in extracting patterns and knowledge from complex datasets. It discusses challenges such as the curse of dimensionality and the importance of dimensionality reduction techniques like feature selection and PCA. It also showcases the use of optimization strategies in visualizing data, whether through supervised or unsupervised methods. Real-world applications, including global behavior modeling for grid autonomic management, highlight the practical implications of these concepts.

tamar
Télécharger la présentation

Understanding Data Mining and Visualization: Techniques and Applications

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Data Mining VS Visualization Santiago González Tortosa <sgonzalez@fi.upm.es>

  2. Contents • Data Mining VS Visualization • Visualizeto DM • DM toVisualize (to DM) • Real worldwork: • Global BehaviorModeling: A New approachtoGridautonomicmanagement

  3. Data Mining VS Visualization • Data Mining • Knowledgediscovery and extration • Notalwaysiseasytoseepatterns, distributions, etc. • Visualization • Represents data (2D, 3D, Virtual Reality,…) • Helpstoextractpatterns • Notalwaysiseasytorepresent data in 2 or 3 dimensions

  4. Visualizeto DM • Visualizationhelpustoextractanypattern in the data

  5. Visualizeto DM • Visualizationhelpustoextractanypattern in the data

  6. DM toVisualize • Data contains N (> 3) features • Curse of Dimensionality • Wewanttovisualizeall data • DimensionalityReduction • Reduce number of features • Transform and create new features

  7. DM toVisualize • DimensionalityReduction • L.J.P. van der Maaten, E.O. Postma, and H.J. van den Herik. DimensionalityReduction: A ComparativeReview. TilburgUniversityTechnicalReport, TiCC-TR 2009-005, 2009 • Convextechniques: optimizeanobjective function that does not contain any local optima • Nonconvextechniques: optimizeobjective functions that do contain local optima

  8. DM toVisualize • Optimizationtechniques (hillclimbing, evolutive, etc.)

  9. DM toVisualize • Optimizationtechniques • Oneobjective • Oneobjectivewithconstraints (Semi-Supervisedand labeling) • Multiobjective

  10. DM toVisualize • Example: Optimize axis

  11. DM toVisualize • DimensionalityReduction in 2 phases: • FSS: FeatureSubsetSelection (wrapper, needed CLASS!) • Transformation and creation of new features (f.e. PCA)

  12. DM toVisualize • Example of DimensionalityReduction in 2 phases • Userexpertinteracts

  13. DM toVisualize • DM toVisualize….to DM!! • The idea istoobtain new knowledgeorpatternsviewingthe data. • Supervisedinfo: data withthesameclass are represented in thesamearea (KNN). • Unsupervisedinfo: data isagrouped

  14. DM toVisualize • Examplethatsome data isagrouped

  15. DM toVisualize • Visualization • 2D and 3D visualization • Virtual Reality • Inmersion • Interaction • Imagination • AugmentedReality

  16. Real worldwork Global BehaviorModeling: A New approachtoGridautonomicmanagement Jesus Montes <jmontes@fi.upm.es>

  17. Data Mining VS Visualization Santiago González Tortosa <sgonzalez@fi.upm.es>

More Related