De Wijkwizard

  • 214 views
Uploaded on

Presentatie van de Wijkwizard (the neighbourhoudwizard) at the DDSS conference 2006

Presentatie van de Wijkwizard (the neighbourhoudwizard) at the DDSS conference 2006

More in: Technology
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
    Be the first to like this
No Downloads

Views

Total Views
214
On Slideshare
0
From Embeds
0
Number of Embeds
1

Actions

Shares
Downloads
1
Comments
0
Likes
0

Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide

Transcript

  • 1. Léon van Berlo / Jos van Leeuwen The Neighbourhood Wizard Cause and effect of changes in urban neighbourhoods
  • 2. Agenda
    • Introduction
    • Objective
    • Approach
    • (Experiencing) Liveability
    • Data Collection
    • Knowledge representation
    • Prototype
    • Evaluation and testing
    • Conclusions and future work
    • Coffee break
  • 3. Introduction
    • Quality of the neighbourhood (physical and social)  Increasingly important
    • Local initiatives for neighbourhood improvement
    • Municipalities support these initiatives  Citizen participation
    • Issues:
      • Inhabitants focus on their own problems (not the ones from their neighbours)
      • Inhabitants don’t see the complex dependencies of a decision
      • Inhabitants give concrete proposals for change in stead of their desire
  • 4. Objective
    • Making citizens realise what the consequences are of their ideas for changes
    • By developing a tool that allows citizens to:
      • propose changes to their neighbourhood;
      • assess the quality of these changes
  • 5. Approach
    • Find a set of indicators for experience of liveability of the neighbourhood
    • Find a set of characteristics that affect the liveability
    • Determine a BBN that represents the knowledge
    • Build a prototype
      • Narrowing its scope to the plaza type of habitat
    • Testing the prototype in the Dutch city of ’s‑Hertogenbosch
  • 6. Experiencing liveability: Leidelmeijer and Marsman 1999
  • 7. Example experience by an individual Appreciation Importance Satisfaction Liveability Characteristics
  • 8. Example experience by an individual
  • 9. Example experience by an individual
  • 10. Experience by another individual
  • 11. Grouping individuals and their needs
    • Wishprofiles:
      • Teenagers
      • Yuppies
      • Families
      • Elderly
      • Handicapped (elderly)
    • Aspects:
      • Space
      • Liveliness
      • Security
      • Quality
      • Status
      • Traffic
  • 12. Data Collection
    • Questionnaire of liveability regarding the city of ’s‑Hertogenbosch
    • Experiences of characteristics such as:
        • ‘public furnishing’
        • ‘available facilities’
        • ‘public accessibility’
        • ‘status’
        • ‘appearance’
        • ‘ambiance’
        • etc.
    • For plazas, over 40 characteristics were included.
    • Scale of seven possible values
      • Ranging from deficient, through moderate and neutral, to ample and excessive.
  • 13. Data Collection Example: Form and function: Incoherent Suitable surprising conflicting
  • 14. Knowledge Representation
    • Bayesian Network:
      • Can deal with uncertainty and interdependent variables
    • Determining the structure of a BN:
      • 1) Knowledge expert who constructs a network
      • 2) Examining data from the particular domain
    • In this project 2 is used to come to a base network which was refined by 1.
  • 15. Structural Learning
    • Hugin (www.hugin.com) was used with:
      • PC algorithm (Peter & Clark)
      • NPC algorithm (Necessary Path Condition)
    • Constraint-based learning algorithms
    • Derive conditional independence and dependence statements by performing statistical tests on pairs of variables in the data set
  • 16. BN Structure (1)
  • 17. Structural Learning
    • PC and NPC  same results
    • Significance level 0.05 – 0.03 – 0.01
    • Difference in ‘real relationships’ and ‘relationships in the data’
    • Defining relations that are not in the data: no use
  • 18. BN Structure (2)
  • 19. Prototype
    • User-interaction focused on a task assigned to the user
      • Users can experience this like a game
    • Representing the effects of changes
    • Representing the desired states of the aspects for different sections of the population
    • Availability of the system on Internet
    • Easy to use interface and obvious navigation
  • 20. Changing elements:
    • Three ways:
      • 1) Drawing
      • 2) Picking from a list
      • 3) Cheating
  • 21. Changing elements: 1 (drawing)
  • 22. Changing elements: 2 (picking from a list)
  • 23. Changing elements: 3 (cheating)
  • 24. Presentation of Predicted Effects
    • Three levels:
      • 1) Simple  does not give desired effect
      • 2) Normal
      • 3) Expert
  • 25. Presentation of Predicted Effects: 2 (normal)
  • 26. Presentation of Predicted Effects: 2 (normal)
  • 27. Presentation of Predicted Effects: 3 (expert)
  • 28. Presentation of Predicted Effects: 3 (expert)
  • 29. Evaluation
    • www.WijkWizard.nl (dutch)
    • Tested and evaluated by inhabitants of the city of ’s‑Hertogenbosch.
    • Online evaluation form.
    • “ Thanks to the Neighbourhood Wizard, I now see that certain ideas are positive for me, but negative for other members of our community” : 7.4
    • “ The Neighbourhood Wizard shows me that changes can have positive effects on one aspect, but negative effects on other aspects” : 7.0
    • Confirmed the educational function of the prototype!
  • 30. Conclusions (+)
    • The Neighbourhood Wizard helps users to see that certain ideas are positive for them, but negative for other sections of the population;
    • The Neighbourhood Wizard shows users that changes can have positive effects on one aspect, but negative effects on other aspects;
    • The Neighbourhood Wizard helps users to realize the complexity of a design task and as a result users will have a better informed view on plan proposals and probably a higher appreciation of plans.
  • 31. Conclusions (-)
    • Design of the user interface
    • Navigation structure (too many clicks)
    • Abstract terms
      • Inclusion of more concrete elements (number of parking lots) can help take away long-living irritations that inhabitants may have
    • The data collection is restricted to physical characteristics
  • 32. Future work
    • Investigate the relations between characteristics in depth
    • (developing a technique that) Includes explanations of the effects
      • In some cases the predictions are not so obvious and require further explanation
      • For example: The creation of a quiet plaza has negative effects on the safety of the plaza. This is not a logical, though correct, prediction because the quietness of a plaza will attract criminal behaviour
  • 33. Thank you
    • Questions or coffee break?
    Leon.vanBerlo@tno.nl / WijkWizard.nl j.p.v.leeuwen@tue.nl / www.ddss.nl