Discovering, Visualizing and Sharing Knowledge through Personalized - - PowerPoint PPT Presentation

▶
discovering visualizing and sharing knowledge through
SMART_READER_LITE
LIVE PREVIEW

Discovering, Visualizing and Sharing Knowledge through Personalized - - PowerPoint PPT Presentation

Nov 06, 2023 24 likes •239 views

Discovering, Visualizing and Sharing Knowledge through Personalized Learning Knowledge Maps The AWAKE Project personal knowledge structure (user A) A m E ?? y map grid semantische ? C d achsen o netz awarenes cs kuns s multi-

slide-1
SLIDE 1

Discovering, Visualizing and Sharing Knowledge through Personalized Learning Knowledge Maps

Jasminko Novak, Michael Wurst

The AWAKE Project

interactiv e multi- user distribute d s t a g e Inte rnet a r t m e d i a l s t a g i n g netz kuns t C S C W Cluster Modification Agents Agent B Agent C Agent D Agent Modifiers Learned User Profiles Delineate Agent E MyAgent Splitter Mixer Cross ? Data Sources COR DIS Data Adapters

D a t a b a s e S e m i a u t
  • n
  • m
d u r c h

map grid semantische achsen Data Entry ? A E C

D a t a b a s e S e m i a u t
  • n
  • m
d u r c h D a t a b a s e S e m i a u t
  • n
  • m
d u r c h

personal knowledge structure (user A) A E C m y d

  • cs

?? ? inter activ e art mixed reality awarenes s virtual reality collabora tive

slide-2
SLIDE 2

2

Context and Challenge

Challenge: Capture, visualize and exchange existing knowledge of user groups Support for heterogeneous expert communities Discovering connections across different domains Exchange of knowledge between individual experts

slide-3
SLIDE 3

3

Practical Approach

Situation: Group of users explores an information space Goal: Capture user knowledge reflected in their interaction with information

Knowledge Maps

1) Creating a context for the interpretation of users’ structuring actions 2) Acquisition and visualization

  • f similarity relations

between users, documents and topics

slide-4
SLIDE 4

4

Agent Architecture

Shared Data Space Feature Extraction Agents Learning Agents Clustering Agent Map Personalization Agent Search Agent Map Editor Topic Agent Personalized Navigation (online) Data Analysis (offline)

Heterogeneous Document Sources

slide-5
SLIDE 5

5

Personal Agents

System generated Knowledge Maps

Video Sound Interactive Show Agents Art

  • System generated Knowledge Maps
  • Detailed document information
  • Related (similar) documents
  • Personal Map Editor
  • Search for Personal Maps
  • Document pool personalization
  • Topic network
slide-6
SLIDE 6

6

Personal Agents

System generated Knowledge Maps

Video Sound Interactive Show Agents Art Document title Author(s) Abstract Keywords ...

  • System generated Knowledge Maps
  • Detailed document information
  • Related (similar) documents
  • Personal Map Editor
  • Search for Personal Maps
  • Document pool personalization
  • Topic network
slide-7
SLIDE 7

7

Personal Agents

System generated Knowledge Maps

Video Sound Interactive Show Agents Art

  • System generated Knowledge Maps
  • Detailed document information
  • Related (similar) documents
  • Personal Map Editor
  • Search for Personal Maps
  • Document pool personalization
  • Topic network
slide-8
SLIDE 8

8

Video New Cluster Sound

Personal Agents

Personal Knowledge Maps

  • System generated Knowledge Maps
  • Detailed document information
  • Related (similar) documents
  • Personal Map Editor
  • Search for Personal Maps
  • Document pool personalization
  • Topic network
slide-9
SLIDE 9

9

Personal Agents

Search for Personal Knowledge Maps

New Cluster Sound Video New Cluster Sound Video New Cluster Sound

  • System generated Knowledge Maps
  • Detailed document information
  • Related (similar) documents
  • Personal Map Editor
  • Search for Personal Maps
  • Document pool personalization
  • Topic network
slide-10
SLIDE 10

10

Learn&Apply to Document Pool

Video New Cluster Sound

Personal Agents

Personalized Knowledge Maps

  • System generated Knowledge Maps
  • Detailed document information
  • Related (similar) documents
  • Personal Map Editor
  • Search for Personal Maps
  • Document pool personalization
  • Topic network
slide-11
SLIDE 11

11

distributed theatre virtual multi-user interactive

Personal Agents

Creating a Collaborative Topic Net

  • System generated Knowledge Maps
  • Detailed document information
  • Related (similar) documents
  • Personal Map Editor
  • Search for Personal Maps
  • Document pool personalization
  • Topic network
slide-12
SLIDE 12

12

Personal Agents

slide-13
SLIDE 13

13

Agent Architecture

Shared Data Space Feature Extraction Agents Learning Agents Clustering Agent Map Personalization Agent Search Agent Map Editor Topic Agent Personalized Navigation (online) Data Analysis (offline) Heterogeneous Document Sources

slide-14
SLIDE 14

14

Shared Data Space

Extracted Meta Information

Feature Extraction Agents Learning Agents Personal Agents Visualization Client

Relationships between documents, topics and users

documents topics users Shared Data Space Raw data

slide-15
SLIDE 15

15

Content and Context Analysis

Combined Similarity Measure

Learning (similarity) relations between documents, topics and us ers

Learned relationships

Text Analysis Agent Reference Analysis Agent content

slide-16
SLIDE 16

16

Content and Context Analysis

Combined Similarity Measure

Learning (similarity) relations between documents, topics and us ers

Learned relationships

Text Analysis Agent Reference Analysis Agent content context Context Analysis Agents

slide-17
SLIDE 17

17

0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 1 3 5 7 9 11 13 15 17 19 Number of maps

  • avg. err.

Text Context Combination

Content and Context Analysis

slide-18
SLIDE 18

18

Some Relevant questions from a Machine Learning Viewpoint

  • Amount of user interaction dataneeded to learn the

desired concepts?

  • Robustness of the extracted knowledge against noise or

malicious manipulation

  • Local/Global Pattern (e.g. how are minor opinions

reflected in the resulting structures?)

Content and Context Analysis

slide-19
SLIDE 19

19

Application Domain

netzspannung.org

  • Knowledge portal connecting media

art, design and technology

  • Collaborative information pool

(projects, events...)

  • Very heterogeneous content
  • different categorization schemes
  • constantly growing
slide-20
SLIDE 20

20

Summary

  • Model and prototype for capturing, visualizing and exchanging

knowledge

  • Agent technology as software architecture and data

integration model

  • Machine Learning methods for combining context and content

data

Summary and Future Work

Future work

  • Evaluation and user testing
  • Improvement of visualization and data analysis methods
  • Support for the Topic Map Format
slide-21
SLIDE 21

21

Project & Partners

Project partners

Fraunhofer Institute for Media Communication

MARS Exploratory Media Lab (Project Leader)

Fraunhofer Institute for Industrial Engineering

Competence Center Softwaretechnologie & interactive Systems University of Dortmund Artificial Intelligence Unit University of Siegen

  • Dept. for Parallel Systems

Thank you for your attention!