Your SlideShare is downloading. ×
4A_1_Uncertainty in the 2001 output area classification for the census of england and wales
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×

Saving this for later?

Get the SlideShare app to save on your phone or tablet. Read anywhere, anytime - even offline.

Text the download link to your phone

Standard text messaging rates apply

4A_1_Uncertainty in the 2001 output area classification for the census of england and wales

517
views

Published on


0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total Views
517
On Slideshare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
0
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. Uncertainty in the 2001 Output Area Classification for the Census of England and Wales
    Peter Fisher
    Department of Geography, University of Leicester
    GISRUK , University College,
    London, 15th April 2010
  • 2. Outline
    Output area classification
    Uncertainty reporting
    Transforming uncertainty
    Fuzzy c Means
    Possibilistic c Means
    Results
    Conclusion
  • 3. Output Area Classification
    ONS OAC
    Hard Classification
    Based on 41 census variables
    Using the k Means classification
    Three levels recognised
    Super-Groups – 7 classes
    Groups – 21 classes
    Sub-Groups – 52 classes
  • 4. Super-Groups
    Blue Collar Communities
    City Living
    Countryside
    Prospering Suburbs
    Constrained by Circumstances
    Typical Traits
    Multicutural
  • 5. Each is characterised by variables
    Blue Collar
    Multicultural
    Typical Traits
    From Vickers, Rees and Birkin, (2005) WP 05/2 School of Geography, University of Leeds
  • 6. Uncertainty
    Hardly anywhere could be a perfect example of any such class
    How many cultures are required for an area to be multicultural?
    How many white collar (or pale blue collar) workers are allowed in a blue collar community?
    Can you have prosperous suburbs in the country or in the city?
    What are “Typical traits”?
    Similar questions could be asked of other classifications
  • 7. Uncertainty Reporting
    Uniquely among geodemographic classifications ONS OAC reports uncertainty
    Distance to cluster centres is reported for ALL classes at all levels of classification
    These distances could be derived from:
    the original data and the cluster centroids
  • 8. Fuzzy c Means
    Fuzzy membership is given by:
    Subject to the conditions that:
    for all i and j
    for all i, and
    for all i.
  • 9. Possibilistic c Means
    Relax condition 3 to simply
    for all j.
    Using
  • 10. Where …
    m is the fuzziness as in FCM
    ηi is found iteratively from
    It can be different for each class
  • 11. Study area: Leicester
  • 12.
  • 13.
  • 14. Entropy – Degree of confusion
    FCM 0-1
    Single class to more classes
    PCM 0-<c
  • 15. Confusion
    The same number of cases in each OA Class as in the actual OAC
    Different α-cut for each class
  • 16. PCM
    FCM
  • 17. Conclusion
    Possibilistic c-Means offers a alternative to the usual Fuzzy c-Means
    Abandons constrain of memberships summing to 1
    Makes more sense ?
    But the total class areas do not sum to 100% of the study area (usually greater)?
    Treatment of the uncertainty offers more satisfying (?) outcomes of the OAC
    But this treatment is only possible for the OAC
  • 18. Questions ?
    Email: pff1@le.ac.uk