Your SlideShare is downloading. ×
0
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
Trajectory Mining Icdm 2009
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

Trajectory Mining Icdm 2009

773

Published on

SSTDM-ICDM 2009

SSTDM-ICDM 2009

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

  • Be the first to like this

No Downloads
Views
Total Views
773
On Slideshare
0
From Embeds
0
Number of Embeds
1
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
  • Morning everyoneNow I would like to give a brief introduction about my previous work, actually it`s my first time to attend this workshop and ICDM.So if you have any questions during my talk, pls feel free to ask meOk, then So our motivation of this work is to analyze ...The result of this research could be apply to …Next slide we will give two picture to explain this motivation more vividly
  • Let us take a look at the left picture , This is a part view of beijing planet map. The red line means the hot tourist path at this coarse level, Of course it dosent come from real data, we draw the line here just want to show our idea.At this level, one centimeter is equal to half a kilometer. Suppose we get the red line by clustering trajectories with some density measure, If we want to look more details of the Forbidden City, we should zoom in as much as is needed.Then we could get the right picture, one centimeter in the picture is equal to two hundred meters, we think the clusters density should be consistent with this grain size , with more low density measure we could get the cluster as this red line
  • The trajectory we mentioned
  • Let us give an illustration of the Frechet distance: a man is walking witha dog on a leash. This man is walking on the one curve, the dog on theother one. Both may vary their speed, but backtracking is not allowed.Then the Frechet distance of the curves is the minimal length of a leashthat is necessary. The Frechet method has the advantage of computing distancesonly on homologous points and not between closest points as for theHausdorff distance
  • The method at this time is not up to do this fast , the problem is to do this in a meaningful way
  • Transcript

    • 1. Multi-granularity Visualization of Trajectory Clusters using Sub-trajectory Clustering<br />Cheng Chang, Bao-Yao Zhou<br />HP Labs China<br />12-6-2009<br />
    • 2. Motivations<br />Analyzing objects movement to identify frequently moving paths at different scale can improve a large set of services such as:<br />Traffic management.<br />Location-based service.<br />Customer behavior analysis.<br />CCTV Surveillance.<br />2<br />15 March 2010<br />
    • 3. Motivations<br />3<br />15 March 2010<br />
    • 4. How to realize <br /> Using Trajectory segment as the main interest unit <br /> Clustering all the Trajectory segment based on density control<br /> Extract the cluster multi-granularity <br />4<br />15 March 2010<br />
    • 5. Trajectory definition<br />Definition 1.A trajectory is a sequence of multi-dimensional points.<br />Definition 2.A cluster is a set of similar sub-trajectories.<br />5<br />15 March 2010<br />
    • 6. Method<br />6<br />Select the set of interesting trajectories<br />Corner detection<br />Segment the trajectories<br />Get the augmented order of the sub-trajectories<br />Frechet distance measure<br />Extract the cluster with different granularity<br />Visualization<br />15 March 2010<br />
    • 7. Trajectory segmentation<br />Scans the point sequence of a trajectory and selects candidate corner points according to the calculation of the open angle <br />Remove redundant candidates. <br />7<br />15 March 2010<br />
    • 8. Distance measure for sub-trajectory clustering<br />Fréchet Distance <br />The maximal distance between two oriented lines.<br />Idea<br /><ul><li>Examine all possible walks.
    • 9. Yields a set M of maximum leash lengths.
    • 10. dF= shortest leash length in M.</li></ul>8<br />15 March 2010<br />
    • 11. Distance measure for sub-trajectory clustering<br />Fréchet Distance<br /> Discrete Fréchet Distance<br />9<br />15 March 2010<br />
    • 12. Density-based clustering<br />A cluster: a maximal set of density-connected points<br />Discover clusters of arbitrary shape in spatial databases with noise<br />OPTICS: ordering points to identify the clustering structure<br />It computes an augmented clustering-ordering for automatic cluster analysis. <br />Based on this idea, two values need to be introduced: <br />Core distance <br />Reachability distance<br />10<br />15 March 2010<br />
    • 13. Density-based clustering<br />OPTICS Algorithm<br />create an ordering of a database, additionally <br />store the core-distance and a suitable reachability-distance for each points. <br />Then such information is used to extract all density-based clusters with respect to any distance ε ’ smaller than the generating distance ε from this order. <br />11<br />15 March 2010<br />
    • 14. Extract sub-trajectory clusters<br />Scanning the order and assigning the cluster-membership<br />Check the ascription of each sub-trajectory in cluster<br />12<br />15 March 2010<br />
    • 15. Density-based clustering<br />OPTICS Motivation<br />input parameters (e.g., Eps) are difficult to be determined <br /> We let the user to explore the clustering result at different scale<br />one global parameter setting may not fit all the clusters<br />Through the visualization<br />it’s good to allow users to have flexibility in selecting clusters<br />Yes!<br />13<br />15 March 2010<br />
    • 16. Visualization<br />14<br />15 March 2010<br />
    • 17. Experiments<br />The experiments are performed on the animal movement dataset provided by the Starkey project<br />15<br />15 March 2010<br />
    • 18. Experiments<br />16<br />15 March 2010<br />
    • 19. Conclusion<br />A new framework to discover multi-granularity clusters by clustering sub-trajectories. <br />An extension of the OPTIC algorithm for curve segments by using Fréchet distance measurement.<br />The user can tune the density to explore the visualization of clustering results at different spatial granularities.<br />17<br />15 March 2010<br />
    • 20. Thank you for your attention.<br />Questions?<br />18<br />15 March 2010<br />

    ×