Successfully reported this slideshow.

Finding Patterns in Temporal Data

3

Share

FINDING PATTERNS
IN TEMPORAL DATA
                                                   osium!
                              ...
FINDING PATTERNS
IN TEMPORAL DATA
                                                   osium!
                              ...
TEMPORAL CATEGORICAL DATA
•  A type of time series
                              Event!           Category!         Event!...

YouTube videos are no longer supported on SlideShare

View original on YouTube

YouTube videos are no longer supported on SlideShare

View original on YouTube

1 of 32
1 of 32

More Related Content

More from Krist Wongsuphasawat

Related Books

Free with a 14 day trial from Scribd

See all

Finding Patterns in Temporal Data

  1. 1. FINDING PATTERNS IN TEMPORAL DATA osium! CI L Symp KRIST WONGSUPHASAWAT 27th H 010! , 2 May 27 TAOWEI DAVID WANG CATHERINE PLAISANT BEN SHNEIDERMAN HUMAN-COMPUTER INTERACTION LAB UNIVERSITY OF MARYLAND
  2. 2. FINDING PATTERNS IN TEMPORAL DATA osium! CI L Symp KRIST WONGSUPHASAWAT 27th H 010! , 2 May 27 TAOWEI DAVID WANG CATHERINE PLAISANT BEN SHNEIDERMAN HUMAN-COMPUTER INTERACTION LAB UNIVERSITY OF MARYLAND
  3. 3. TEMPORAL CATEGORICAL DATA •  A type of time series Event! Category! Event! Numerical! Stock: Microsoft Patient ID: 45851737 $'#")#"$!$&!$($$ &0!B$0& !"#$"#"$$%&!'(") &*++,-./& $'#")#"$!$&!$(!; &0!B$!& !"#$"#"$$%&!'(0) &123+43567& $'#")#"$!$&!$(0$ &0!B$"& !"#$"#"$$%&""('' &89:& $'#")#"$!$&!$('; &0!B$%& !"#$;#"$$%&$;($< &=/>>+& $'#")#"$!$&!!($$ &0!B!)& !"#!'#"$$%&$)(!? &1@,A& & Time Arrival Emergency ICU Floor Exit
  4. 4. TEMPORAL CATEGORICAL DATA Electronic Health Records: symptoms, treatment, lab test" Traffic incident logs: arrival/departure time of each unit" Student records: course, paper, proposal, defense, etc." " Others: web logs, usability study logs, etc."
  5. 5. 10+ years work on temporal visualization (mostly on Electronic Health Records)
  6. 6. LIFELINES D SINGLE RECOR [Plaisant et al. 1998] http://www.cs.umd.edu/hcil/lifelines
  7. 7. LifeLines – Single Patient
  8. 8. working with physicians at WASHINGTON HOSPITAL CENTER
  9. 9. EXAMPLE DATA •  Patient transfers ARRIVAL Arrive the hospital EMERGENCY Emergency room ICU Intensive Care Unit INTERMEDIATE Intermediate Medical Care FLOOR Normal room EXIT-ALIVE Leave the hospital alive EXIT-DEAD Leave the hospital dead
  10. 10. TASKS •  Example: Finding “Bounce backs” ICU Floor ICU within 2 days
  11. 11. LIFELINES 2 RECORD D R ECOR RECORD RECORD RECORD [Wang et al. 2008, 2009] http://www.cs.umd.edu/hcil/lifelines2
  12. 12. Multiple Records ARF (Align-Rank-Filter) Framework Temporal Summary LifeLines2 – Search and Visualize
  13. 13. ALIGNMENT •  Sentinel events as reference points Time June July August Patient #45851737 Arrival Emergency ICU Floor Exit Patient #43244997 Arrival Emergency ICU Floor Exit
  14. 14. ALIGNMENT (2) •  Time shifting Time 0 1M 2M Patient #45851737 Admit Emergency ICU Floor Exit Patient #43244997 Admit Emergency ICU Floor Exit
  15. 15. SIMILAN RECORD D R ECOR RECORD RECORD RECORD [Wongsuphasawat & Shneiderman 2009] http://www.cs.umd.edu/hcil/similan
  16. 16. Similan – Search by Similarity
  17. 17. Similan – Search by Similarity
  18. 18. FINDING “BOUNCE BACKS” Before After •  Much faster to specify new query •  Visualizing the results gives better understanding
  19. 19. USER STUDIES: SEARCH LifeLines2 Similan Exact! Similarity-based! MUST have A, B, C SHOULD have A, B, C Query Query Record#2 Record#2 more Record#1 Record#1 similar Record#3 Record#3
  20. 20. USER STUDIES: SEARCH LifeLines2 Similan Exact! Similarity-based! MUST have A, B, C SHOULD have A, B, C Query Query 1 Record#2 Record#2 more Record#1 Record#1 similar Record#3 Record#3
  21. 21. NEW STUFF Needs for an overview -> LifeFlow!!
  22. 22. TASKS •  Example: Finding “Bounce backs” ICU Floor ICU within 2 days •  Other questions Arrival ? ICU ? ?
  23. 23. LIFEFLOW RECORD D VISUALIZE R ECOR RECO Display the aggregation RD RECORD RECORD RECORD RECORD RECORD AGGREGATE Merge multiple records into tree
  24. 24. AGGREGATE •  Aggregate by prefix #1 #2 #3 #4 Example with 4 records
  25. 25. AGGREGATE •  Aggregate by prefix #1 #2 #3 #4
  26. 26. VISUALIZE •  Inspired by the Icicle tree [Fekete 2004] Number of files!
  27. 27. VISUALIZE (2) •  Use horizontal axis to represent time •  Video
  28. 28. DEMO – LIFEFLOW When the lines are combined into flow!
  29. 29. FUTURE WORK •  Comparison ICU ICU Intermediate Intermediate Jan-Mar 2008 April-June 2008 Floor
  30. 30. TAKE-AWAY MESSAGE Information visualization is a powerful way to explore temporal patterns. You can work with us on new case studies.

×