• No results found

Data Visualization and Data Mining

N/A
N/A
Protected

Academic year: 2021

Share "Data Visualization and Data Mining"

Copied!
35
0
0

Loading.... (view fulltext now)

Full text

(1)

Data Visualization and Data Mining

Data Visualization and Data Mining

Jürgen

Jürgen Symanzik Symanzik

Utah State University, Logan, UT Utah State University, Logan, UT

*e-mail:

*e-mail: symanzik@ [email protected] WWW: http://www.math.

(2)

Contents

Contents

n

n

Data Visualization / Data Mining

Data Visualization / Data Mining

– Software & ToolsSoftware & Tools –

– ApplicationsApplications

n

n

Immersive

Immersive

Projection Technology

Projection Technology

– CAVE & C2CAVE & C2 –

(3)

Visual Data Mining

Visual Data Mining

n

n

Working Definition:

Working Definition:

– Find structure (cluster, unusual observations) inFind structure (cluster, unusual observations) in large and not necessarily homogeneous

large and not necessarily homogeneous data sets using graphical methods and data sets using graphical methods and user interaction

user interaction –

– Goal or expected outcome of explorationGoal or expected outcome of exploration usually unknown in advance

(4)

The Software:

The Software:

XGobi

XGobi

Swayne

Swayne, Cook and , Cook and Buja Buja

• Interactive environment for exploring multivariate dataInteractive environment for exploring multivariate data *

*Linked views allow Linked views allow ``linked brushing’’``linked brushing’’ *

*UnivariateUnivariate, , Bivariate Bivariate and Multivariate views of the dataand Multivariate views of the data *

*Grand tourGrand tour *

*Wide variety of methodsWide variety of methods *

*Open sourceOpen source *

*FreeFree •

• CaveatsCaveats *

(5)

The Software:

The Software:

ExplorN

ExplorN

Carr,

Carr, WegmanWegman, , Luo Luo

• Interactive environment for exploring Interactive environment for exploring multivariate data (similar to

multivariate data (similar to XGobiXGobi)) *

*Advanced Parallel Advanced Parallel CoordinatesCoordinates Displays Displays *

*3D Surfaces3D Surfaces *

*Stereoscopic DisplaysStereoscopic Displays •

• CaveatsCaveats *

*Only on SGI platformsOnly on SGI platforms *

(6)

The Software:

The Software:

ArcView

ArcView

ESRI

ESRITMTM •

• Desktop Desktop GIS GIS with wide range of viewingwith wide range of viewing and data manipulation functions

and data manipulation functions

*

*Editing featuresEditing features *

*Query operationsQuery operations *

*Map displayMap display *

*Interactive interfaceInteractive interface *

*High level internal scripting languageHigh level internal scripting language •

• ArcView ArcView has a wide user basehas a wide user base •

• CaveatsCaveats *

(7)

The Software:

The Software:

XploRe

XploRe

Härdle

Härdle, , KlinkeKlinke, et al., et al.

• Statistical Computing EnvironmentStatistical Computing Environment *

*Built in set of statistical methodsBuilt in set of statistical methods *

*High level programming languageHigh level programming language *

*Highly Interactive GraphicsHighly Interactive Graphics *

*Built in editors for data and programsBuilt in editors for data and programs *

*High level help systemHigh level help system •

• CaveatsCaveats •

(8)

The

The

ArcView

ArcView

/

/

XGobi

XGobi

/

/

XploRe

XploRe

Link

Link

n

n

Links these 3 packages using

Links these 3 packages using

RPCs

RPCs

n

n

Remote Procedure Calls (RPCs

Remote Procedure Calls (

RPCs):

):

– Process on the local system (client) invokes aProcess on the local system (client) invokes a procedure on a remote system (server)

procedure on a remote system (server) –

– Request = client’s desire to execute a particularRequest = client’s desire to execute a particular remote procedure

remote procedure –

– Response = result produced by the remoteResponse = result produced by the remote procedure

procedure

n

(9)

Tools: Linked Brushing 1

Tools: Linked Brushing 1

XGobi

(10)

Tools: Linked Brushing 2

Tools: Linked Brushing 2

(11)

Tools: Parallel Coordinate Plots

Tools: Parallel Coordinate Plots

(12)

Tools:

Tools:

Scatterplot

Scatterplot

Matrix

Matrix

(13)

Applications

Applications

n

n

Environmental Data

Environmental Data

n

n

Remote Sensing

Remote Sensing

n

n

Archaeological Data

Archaeological Data

n

(14)

Environmental Data

Environmental Data

XGobi

(15)

Remote Sensing

Remote Sensing

(16)

Archaeological Data 1

Archaeological Data 1

(17)

Archaeological Data 2

Archaeological Data 2

(18)

Human Motion Data 1

Human Motion Data 1

(19)

Human Motion Data 2

Human Motion Data 2

(20)

IPT / CAVE /

(21)

CAVE Concept

CAVE Concept

n

n

A Projection-based Immersive

A Projection-based

Immersive VR System

VR System

– Silicon Graphics-based with 8 to 12 processorsSilicon Graphics-based with 8 to 12 processors –

– RERE2 2 or RE Infinity graphics enginesor RE Infinity graphics engines

– CRT-based projection systemCRT-based projection system –

– StereographicsStereographics Crystal Eyes shutter glasses Crystal Eyes shutter glasses –

– Head trackingHead tracking –

– Usually 3 to 5 wall cubeUsually 3 to 5 wall cube

n

n

Developed originally at U. Illinois, Chicago

Developed originally at U. Illinois, Chicago

(22)

Statistics in the C2 - 1

Statistics in the C2 - 1

(23)

Statistics in the C2 - 2

Statistics in the C2 - 2

(24)

Statistics in the C2 - 3

Statistics in the C2 - 3

(25)

CAVE Strengths

CAVE Strengths

n

n

Effective immersive

Effective

immersive environment

environment

– Lightweight non-intrusive glassesLightweight non-intrusive glasses –

– Can see own hands and other participantsCan see own hands and other participants

n

n

Effective for group VR

Effective for group VR

(26)

CAVE Weaknesses 1

CAVE Weaknesses 1

n

n

CRT Projectors

CRT Projectors

– Projectors not very brightProjectors not very bright –

– Shock, vibration & heat, hard to keep focusShock, vibration & heat, hard to keep focus –

– Geometric distortion at wall interfacesGeometric distortion at wall interfaces

n

n

Tracking

Tracking

– One user tracked, badly distorted stereo forOne user tracked, badly distorted stereo for users not at viewpoint

users not at viewpoint

n

n

User Interface

User Interface

(27)

CAVE Weaknesses 2

CAVE Weaknesses 2

n

n

Expensive

Expensive

– $1,000,000 fully outfitted$1,000,000 fully outfitted –

– $600,000+ SGI computers$600,000+ SGI computers –

(28)

Motivation for

Motivation for

MiniCAVE

MiniCAVE

n

n Installed MATLAB 5 on SGI Onyx and PentiumInstalled MATLAB 5 on SGI Onyx and Pentium

– Benchmarks on 200 megahertz Pentium Pro ($3000)Benchmarks on 200 megahertz Pentium Pro ($3000) and 200 megahertz SGI Onyx ($120,000) similar

and 200 megahertz SGI Onyx ($120,000) similar

n

n Liquid Crystal Projectors sharp, bright, and stableLiquid Crystal Projectors sharp, bright, and stable under shock, vibration and temperature variation under shock, vibration and temperature variation

n

n StereographicsStereographics Crystal Eyes technology available Crystal Eyes technology available for Windows NT

(29)

MiniCAVE

MiniCAVE

Concept

Concept

n

n

Windows NT/Intel Pentium II 400 mhz

Windows NT/Intel Pentium II 400

mhz

n

n

LCD-based projection systems

LCD-based projection systems

n

n

12 ft cubes scaled to 6 ft cubes

12 ft cubes scaled to 6 ft cubes

n

n

Tracking optional, reduced latency

Tracking optional, reduced latency

n

n

Voice command metaphor

Voice command metaphor

n

(30)

n

n

Stereo using CrystalEyes

Stereo using

CrystalEyes

– Above-below stereoAbove-below stereo –

– Image resolution 1024x384 each eyeImage resolution 1024x384 each eye –

– Vertical refresh rate 120-150 (60-75 each eye)Vertical refresh rate 120-150 (60-75 each eye)

n

n SGI monitor can handle both 120 and 150SGI monitor can handle both 120 and 150

n

n CRT projector can only handle 120 refresh rateCRT projector can only handle 120 refresh rate

Implementation Using Monitor

Implementation Using Monitor

(31)

Principles of Above-Below

Principles of Above-Below

CrystalEyes

CrystalEyes

Stereo

Stereo

Graphics Card

Computer

Video signal

CrystalEye Emitter

Double Sync video signal

Monitor/Projector

Emitted pulse

Left Right

(32)

Speech Motivation

Speech Motivation

n

n

User Interfaces (Van Dam)

User Interfaces (Van Dam)

n

n

Shortcuts in

Shortcuts in

XGobi

XGobi

n

(33)

MiniCAVE

MiniCAVE

- Successes

- Successes

n

n

Port of SkyFly

Port of

SkyFly Stereoscopic Demo to NT

Stereoscopic Demo to NT

successful with adequate frame rates on 333

successful with adequate frame rates on 333

megahertz machine

megahertz machine

n

n

CrystalEyes interface on NT successful

CrystalEyes

interface on NT successful

n

n

Speech recognition using Dragon Dictate

Speech recognition using Dragon Dictate

successful

successful

(34)

MiniCAVE

MiniCAVE

- Future Steps

- Future Steps

n

n

NSF proposal submitted - waiting for

NSF proposal submitted - waiting for

funding of required hardware

funding of required hardware

n

(35)

Contact

Contact

n

n

Jürgen Symanzik

Jürgen

Symanzik

– symanziksymanzik@[email protected]

n

n

Website

Website

References

Related documents

In Nigeria today, there is a growing concern for food security with particularly focus on agricultural products due to increasing population (Chen, 2003). With income as a

Finally, I conducted additional research in authentication methods as part of six large research and development projects and served as the research advisor to 29 informatics

property, it is often the desire of the shareholders to transfer the leased ground (and perhaps some equipment) to the newly created corporation with the intention that the departing

Specifically, passive participants were shown all the screenshots given to active participants in Stage I and Stage II, and, after being informed about whether that active

This study reports on detection, localization, and quanti fication of frequent small rockfalls and infrequent pyroclastic density currents descending the southeast flanks of

He used social network analysis, text mining and correlation analysis to compare the user behavior in two similar local campus-based social media services in Finland and China from

Potential explanations for the large and seemingly random price variation are: (i) different cost pricing methods used by hospitals, (ii) uncertainty due to frequent changes in

Players can create characters and participate in any adventure allowed as a part of the D&D Adventurers League.. As they adventure, players track their characters’