digital data visualization may 1, 2001 hwan-seung yong dept. of computer science & eng ewha...

Digital Data Visualization

May 1, 2001

Hwan-Seung Yong

Dept. of Computer Science & Eng

Ewha Womans Univ.

[email protected]

Data Visualization 2

Contents

• Background

• Visualization Example– OLAP

– Data Mining• Multimedia Data Mining

• Spatial Data Mining

• Text Mining

• New Visual Approach – Visual ICON Language

– Visual Language

• Future Trend


Database View of Data Visualization

• File Processing: – Record by record navigation,

• Network/Hierarchical Data Model– Record based interface using Text

– Records have network/hierarchical structure

• Conceptual Modeling, Database Design– Entity-Relationship Model

– ER Diagram

• Relational Model– 2 dimensional Table

– QBE User Interface: 2 dimensional


ERwin Database Designer


Access Query Interface: QBE


Definition of Visualization

• To form a mental vision, image, or picture of something not visible (an abstraction)– To make visible to the mind or imagination

– [Oxford Dictionary, 1989]

• Visualization is a method of Computing– It transforms the symbolic into geometric, enabling researchers to

observe their simulations and computation.

– Enrich the process of scientific discovery

– Foster profound and unexpected insights

– In many fields, it is already revolutionizing the way scientists do science

– [MCC89]


Scientific Visualization/Goals

• exploration/exploitation of data and information

• enhancing understanding of concepts and processes

• gaining new (unexpected, profound) insights

• making invisible visible

• effective presentation of significant features

• quality control of simulations, measurements

• increasing scientific productivity

• medium of communication/collaboration


Visualization and adjacent disciplines

• Computer Graphics: Efficiency of algorithms (CG) vs effectiveness of use (V).

• Computer Vision: Mapping from pictures to abstract description (CV) vs mapping from abstract description to pictures (V).

• Image Processing: Mapping from data domain to data domain (IP) vs mapping from data domain to picture domain (V).

• Art and Design: Aesthetics and style (AD) versus expressiveness and effectiveness (V).


Kind of Digital Data

• Atomic Value (Numeric, String, Boolean)

• Multimedia Data– Sound & Audio, Video, Text

• Complex Data Structure– Tuple, Set, Array, Stack, Queue, Tree, Graph etc

• Large Set of Data– Database

• What to visualize?

• Why

• How


New Data Processing Technique

• Object-Oriented/Relational Data Model– Complex Data: Graph style

• Multimedia: – Visual Interface is required

– Time/Space/Sound and 3 dimension

• Data Warehousing, OLAP and– Multi-dimensional Modeling and Cube Browser

• Data Mining– Visual Interface for Mining

– Visual data mining• Data pattern analysis

• Clustering


Why Visualization?

• Development of H/W and S/W– Computer graphic and visualization technology

• Interactive and Windows Age

• Visual programming Language– Visual Basic, Visual C++ etc.

• Visual ICON language – Emoticon

• Multimedia and Animation


Scientific Data Visualization


Boxplot Analysis

• Five-number summary of a distribution:Minimum, Q1, M, Q3, Maximum

• Boxplot– Data is represented with a box

– The ends of the box are at the first and third quartiles, i.e., the height of the box is IRQ

– The median is marked by a line within the box

– Whiskers: two lines outside the box extend to Minimum and Maximum


A Boxplot

A boxplot


Visualization of Data Dispersion: Boxplot Analysis


Data Visualization Systems

• AVS, IBM Visualization Data Explorer, SGI Explorer

• Khoros, SciAn, other PD vis packages

• NetMap

• S-Plus, SPSS, MatLab, Mathematica, MAPLE

• XmdvTool, Xgobi

• Xsauci


From Tables and Spreadsheets to Data Cubes

• A data warehouse is based on a multidimensional data model which vie

ws data in the form of a data cube

• A data cube, such as sales, allows data to be modeled and viewed in multi

ple dimensions

– Dimension tables, such as item (item_name, brand, type), or time(day, wee

k, month, quarter, year)

– Fact table contains measures (such as dollars_sold) and keys to each of the rel

ated dimension tables

• In data warehousing literature, an n-D base cube is called a base cuboid. T

he top most 0-D cuboid, which holds the highest-level of summarization, i

s called the apex cuboid. The lattice of cuboids forms a data cube.


Visualization of OLAP Model using Star Schema

time_keydayday_of_the_weekmonthquarteryear

time

location_keystreetcityprovince_or_streetcountry

location

Sales Fact Table

time_key

item_key

branch_key

location_key

units_sold

dollars_sold

avg_sales

Measures

item_keyitem_namebrandtypesupplier_type

item

branch_keybranch_namebranch_type

branch


A Concept Hierarchy: Dimension (location)

all

Europe North_America

MexicoCanadaSpainGermany

Vancouver

M. WindL. Chan

...

......

... ...

...

all

region

office

country

TorontoFrankfurtcity


View of Warehouses and Hierarchies

Specification of hierarchies

• Schema hierarchy

day < {month < quarter; week} < year

• Set_grouping hierarchy

{1..10} < inexpensive


Multidimensional Data

• Sales volume as a function of product, month, and region

Pro

duct

Regio

n

Month

Dimensions: Product, Location, TimeHierarchical summarization paths

Industry Region Year

Category Country Quarter

Product City Month Week

Office Day


A Star-Net Query Model

Shipping Method

AIR-EXPRESS

TRUCKORDER

Customer Orders

CONTRACTS

Customer

Product

PRODUCT GROUP

PRODUCT LINE

PRODUCT ITEM

SALES PERSON

DISTRICT

DIVISION

OrganizationPromotion

CITY

COUNTRY

REGION

Location

DAILYQTRLYANNUALYTime

Each circle is called a footprint


OLAP User Interface: Drilling Down• Drilling Down to the lowest level of Customer Dimension


Examples: Discovery-Driven Data Cubes


Browsing a Data Cube

• Visualization

• OLAP capabilities

• Interactive manipulation


OLAP (Summarization) Display Using MS/Excel 2000


3D Cube Browser


Data Mining Result Visualization

• Presentation of the results or knowledge obtained from data mining in visual forms

• Examples

– Scatter plots and boxplots (obtained from descriptive data mining)

– Decision trees

– Association rules

– Clusters

– Outliers

– Generalized rules


Visualization of Association


Market-Basket-Analysis (Association)—Ball graph


Display of Association Rules in Rule Plane Form


Display of Decision Tree (Classification Results)


Output: A Decision Tree for “buys_computer”

age?

overcast

student? credit rating?

no yes fairexcellent

<=30 >40

no noyes yes

yes

30..40


Visualization of a decision tree in MineSet 3.0


Display of Clustering (Segmentation) Results


C-BIRD: Content-Based Image Retrieval from Digital libraries

Search

by image colors

by color percentage

by color layout

by texture density

by texture Layout

by object model

by illumination invariance

by keywords


Multi-Dimensional Search in Multimedia Databases Color layout


Color histogram Texture layout

Multi-Dimensional Analysis in Multimedia Databases


Refining or combining searches

Search for “blue sky”(top layout grid is blue)

Search for “blue sky andgreen meadows”(top layout grid is blue and bottom is green)

Search for “airplane in blue sky”(top layout grid is blue and keyword = “airplane”)

Mining Multimedia Databases


Multidimensional Analysis of Multimedia Data

• Multimedia data cube– Design and construction similar to that of traditional data cubes from relational da

ta– Contain additional dimensions and measures for multimedia information, such as

color, texture, and shape• The database does not store images but their descriptors

– Feature descriptor: a set of vectors for each visual characteristic• Color vector: contains the color histogram• MFC (Most Frequent Color) vector: five color centroids• MFO (Most Frequent Orientation) vector: five edge orientation centroids

– Layout descriptor: contains a color layout vector and an edge layout vector


Mining Multimedia Databases in


REDWHITE

BLUE

GIFJPEG

By Format

By Colour

Sum

Cross Tab

REDWHITE

BLUE

Colour

Sum

Group By

Measurement

JPEGGIF Small

Very Large

REDWHITEBLUE

By Colour

By Format & Colour

By Format & Size

By Colour & Size

By FormatBy Size

Sum

The Data Cube and the Sub-Space Measurements

Medium

Large

• Format of image• Duration• Colors• Textures• Keywords• Size• Width• Height• Internet domain of image• Internet domain of parent pages• Image popularity



Spatial Relationships from Layout

property P1 next-to property P2property P1 on-top-of property P2

Different Resolution Hierarchy



Classification in MultiMediaMiner


• Special features:– Need # of occurrences besides Boolean existence, e.g.,

• “Two red square and one blue circle” implies theme “air-show”

– Need spatial relationships

• Blue on top of white squared object is associated with brown bottom

– Need multi-resolution and progressive refinement mining

• It is expensive to explore detailed associations among objects at high resolution

• It is crucial to ensure the completeness of search at multi-resolution space

Mining Associations in Multimedia Data


Text Miner: Feature Extracton example from IBM Intelligent Miner


Visual Data Mining & Data Visualization

• Integration of visualization and data mining– data visualization

– data mining result visualization

– data mining process visualization

– interactive visual data mining

• Visual Data Mining: the process of discovering implicit but useful knowledge from large data sets using visualization techniques

• Data visualization– Data in a database or data warehouse can be viewed

• at different levels of granularity or abstraction

• as different combinations of attributes or dimensions

– Data can be presented in various visual forms


Boxplots from Statsoft: multiple variable combinations


Visualization of data mining results in SAS Enterprise Miner: scatter plots


Visualization of association rules in MineSet 3.0


Visualization of cluster groupings in IBM Intelligent Miner


GeoMiner Visualization Example


Spatial Clustering


Spatial Association• Association Rules

– isa(X, "Golf Course") -> closeto(X, "Man-Made Channel") (61%, 61%). isa(X, "Golf Course") & closeto(X, "Secondary road") -> closeto(X, "Open space") (64%, 78%).


Data Mining Process Visualization

• Presentation of the various processes of data mining in visual forms so that users can see– How the data are extracted

– From which database or data warehouse they are extracted

– How the selected data are cleaned, integrated, preprocessed, and mined

– Which method is selected at data mining

– Where the results are stored

– How they may be viewed


Visualization of Data Mining Processes by Clementine


Interactive Visual Data Mining

• Using visualization tools in the data mining process to help users make smart data mining decisions

• Example

– Display the data distribution in a set of attributes using colored sectors or columns (depending on whether the whole space is represented by either a circle or a set of columns)

– Use the display to which sector should first be selected for classification and where a good split point for this sector may be


Interactive Visual Mining by Perception-Based Classification (PBC)


Visual ICON Language

• Video Annotation Problem– 과거에는 비디오 데이타들이 1 회성으로 사용– 전문가들이 주석을 달아 저장 , 검색– 현대는 반복 재사용 비디오의 시대

• 어떻게 비디오 데이터를 검색할 것인가 ?

• Keyword based approach 의 한계– Do not describe temporal structure of video

– Not semantic representation• ‘dog’ and ‘German shepherd’

– Do not describe relations between descriptions• Only ‘man’, ‘dog’ ‘bite’ not “dog bite man”

– Do not scale, set of new keyword increase


Language for representation of Video content

• ICON Annotation Language, why?– Quick recognition and browsing of annotation

– Accurate and readable

– Global, international use

• Example– 'Arnold, an adult male, wears a jacket'

– ‘scene is located inside a bar in United States of America’

– Character action: full body actions, head actions, arm actions, and leg actions


Language for representation of Video content

• Number of object– single object, two objects, or groups of objects

• Media Timeline Editor– Timeline annotation of Icon sentence

• Icon Space, icon palette, – a utility for constructing and retrieving iconic

sentences


Media Timeline Editor


Icon Space


ICONS used (sample)


MIT Visual Language Project


Some Words: Integration of Text and Visual Icon


Future Trend

• Animated Visualization vs static visualization

• 3D Visualization vs 2D Visualization

• 3D with Animated Visualization

• Cinematic Technique is becoming more and more important for User Interface– Lev Manovich, Professor of UCSD – The language of New Media, 2000, MIT Press

• Find New metaphor – Spiral Curve etc.

digital data visualization may 1, 2001 hwan-seung yong dept. of computer science & eng ewha...

Documents

digital data visualizationmay

dimension data warehousing

data domain ip vs mapping

visual c

picture domain v

pictures v

effectiveness v

image processing