Community%20Information%20Management

1
Community Information Management

• Nikos Sarkas
• Social Information Systems Seminar
• DCS, University of Toronto, Winter 2007

2
Community Information Management

• Community people with
• Shared interests (movies, database research)
• Shared purpose (intranets, governments)
• Community needs to
• Query
• Monitor
• Discover
• information about entities and their relationships

3
An Example

• Questions that could potentially interest the
  database research community
• Is there any interesting connection between two
  researchers X and Y (e.g., sharing the same
  advisor)?
• In which course is this paper cited?
• Find all citations of this paper in the past one
  week on the Web
• What is new in the past 24 hours in the database
  research community?

4
Cimple and DBLife

• Cimple framework
• Joint project between Univ. of Illinois and Univ.
  of Wisconsin
• A software platform for customized CIM systems
• DBLife
• http//dblife.cs.wisc.edu/
• A CIM prototype aimed at the database research
  community

5
Deploying a Cimple Application
6
Technical Challenges

• Structured information extraction
• Extracted structure exploitation
• Extracted structure maintenance
• Mass collaboration
• Uncertainty and provenance

7
Structure Extraction

• Elements
• Community expert
• Seed Web data sources
• ER semantic schema
• Extractors
• Challenges
• Entity disambiguation
• Execution plan optimization (efficiency
  performance)

8
Structure Exploitation

• Services
• Keyword search
• Entity profiling
• Notification
• ER graph browsing
• Community newsletter
• Structured querying
• Temporal queries

9
Structure Maintenance

• ER graph update
• Rebuilding is expensive and removes temporal
  dimension
• Incremental maintenance instead
• Extractor maintenance
• Extractors can break down
• Malfunction detection and repair

10
Mass Collaboration

• Dual social aspects of CIM
• For the community
• From the community
• Leverage users to improve quality
• Persuade, trick or extort them into doing so
• Personalized data spaces
• Reputation incentives
• Payment schemes

11
Personalized Data Spaces

• User voting and tagging paradigm (IMDB, YouTube,
  del.icio.us)
• Allow users to personalize and manage private
  versions of public data
• Learn from private actions to improve public data
  space
• Examples

12
Reputation Incentives

• Allow users to correct mistakes
• Why care?
• Mistakes adversely affect reputation
• Helping earns credits
• Challenges
• User-entity authentication
• Conflict reconciliation

13
Payment Schemes

• It is the user that pays
• In order to access a service, she must answer a
  simple question
• Challenges
• Merge multiple, noisy answers into a single
  answer
• Which questions to ask?

14
Uncertainty

• Uncertainty sources
• Information extraction
• Mass collaboration
• Modeling and reasoning with uncertainty
• Answer ranking based on confidence
• Interactive queries
• Reducing uncertainty
• Mass collaboration
• User feedback
• Provenance information

15
Provenance

• Use provenance to understand and reduce
  uncertainty
• Justify answers
• Display the raw data and the operations on them
  that contribute to the answer
• Allow hypothetical questions
• What does the answer look like under this
  assumption?

16
Conclusions

• CIM is text analytics with extra dimensions
• Temporal evolution
• Community feedback

17
The End

• Questions?
• Ideas?