Krist Wongsuphasawat's Dissertation Proposal Slides: Interactive Exploration of Event Sequences in Temporal Categorical Data
Upcoming SlideShare
Loading in...5
×
 

Krist Wongsuphasawat's Dissertation Proposal Slides: Interactive Exploration of Event Sequences in Temporal Categorical Data

on

  • 522 views

Slides from my proposal ... a few years ago.

Slides from my proposal ... a few years ago.

Statistics

Views

Total Views
522
Views on SlideShare
520
Embed Views
2

Actions

Likes
0
Downloads
2
Comments
0

1 Embed 2

https://www.linkedin.com 2

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

CC Attribution-NonCommercial LicenseCC Attribution-NonCommercial License

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    Krist Wongsuphasawat's Dissertation Proposal Slides: Interactive Exploration of Event Sequences in Temporal Categorical Data Krist Wongsuphasawat's Dissertation Proposal Slides: Interactive Exploration of Event Sequences in Temporal Categorical Data Presentation Transcript

    • INTERACTIVE EXPLORATIONOF EVENT SEQUENCESIN TEMPORAL CATEGORICAL DATAKRIST WONGSUPHASAWATDEPARTMENT OF COMPUTER SCIENCE &HUMAN-COMPUTER INTERACTION LABUNIVERSITY OF MARYLAND al! pos at ion Pro Dissert 2010! , April 30
    • INTERACTIVE EXPLORATIONOF EVENT SEQUENCESIN TEMPORAL CATEGORICAL DATAKRIST WONGSUPHASAWATDEPARTMENT OF COMPUTER SCIENCE &HUMAN-COMPUTER INTERACTION LABUNIVERSITY OF MARYLAND al! pos at ion Pro Dissert 2010! , April 30
    • TEMPORAL CATEGORICAL DATA•  A type of time series Category! Numerical Categorical Event! Stock: Microsoft Patient ID: 45851737 04/26/2010&10:00 &31.03& 12/02/2008&14:26 &Admit& 04/26/2010&10:15 &31.01& 12/02/2008&14:26 &Emergency& 04/26/2010&10:30 &31.02& 12/02/2008&22:44 &ICU& 04/26/2010&10:45 &31.08& 12/05/2008&05:07 &Floor& 04/26/2010&11:00 &31.16& 12/08/2008&10:02 &Floor& 04/26/2010&11:15 &31.15& 12/14/2008&06:19 &Exit& & Time Admit Emergency ICU Floor Exit
    • 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.•  Usability study logs•  Etc.
    • working with physicians atWASHINGTON HOSPITAL CENTER
    • INTERACTIVE EXPLORATIONOF EVENT SEQUENCESIN TEMPORAL CATEGORICAL DATAKRIST WONGSUPHASAWATDEPARTMENT OF COMPUTER SCIENCE &HUMAN-COMPUTER INTERACTION LABUNIVERSITY OF MARYLAND al! pos at ion Pro Dissert 2010! , April 30
    • EVENT SEQUENCES•  Examples: “Bounce backs” ICU Floor ICU within 2 days
    • RESEARCH STATEMENT “my research aims to design effective visualization and interaction techniques to support users in exploring event sequences in temporal categorical data.” •  a flexible temporal search approach•  a novel visualization that provides an overview of multiple records
    • RESEARCH QUESTION#1 PRELIM. + PROPOSEDMOTIVATION WORK CONCLUSION RESEARCH RESEARCH QUESTIONS QUESTION#2 PRELIM. + PROPOSED WORK
    • TRADITIONAL SEARCH MODELKnow exactly what they want Query Data * How to specify the query? How to display the data? SQL, TSQL, GUI, ... How to process the query? How to make the search efficient? Indexing methods, algorithms
    • EXPLORATORY SEARCH MODELUncertain about what theyare looking forKnow exactly what they want Insight, Hypotheses * search Query Data refine, formulate
    • UNCERTAIN Floor ICU Exit-Alive QUERY within 2 daysFind somethinguseful and display RESULTS Frustrated! found 0 record
    • EXPLORATORY SEARCH MODELUncertain about what theyare looking forKnow exactly what they want Insight, Hypotheses * search Query Data refine, formulate How to display the data? Difficult to understand the event sequences in the data quickly
    • NEEDS A STARTING POINT Show overview or summaryWhere should I start?
    • RESEARCH QUESTIONS1.  How to support the users when they are uncertain about what they are looking for?2.  How to provide an overview of event sequences for temporal categorical data?
    • RESEARCH QUESTIONS1.  How to support the users when they are uncertain about what they are looking for?2.  How to provide an overview of event sequences for temporal categorical data?
    • APPROACHESExact Search! Similarity-based Search!MUST have A, B, C SHOULD have A, B, C Query Query Record#1 Record#2 more Record#2 Record#1 similar Record#3 Record#3
    • SIMILARITY-BASED SEARCH How to specify query?What is “similar”? How to display results? User Interface!Similarity Measure & Visualization
    • SIMILARITY MEASUREGives a score that shows how much a record is similar to the query. Min = 0 / Max = 1 Query Record#1 0.80 Record#2 0.63 Record#3 0.96 Record#4 0.77
    • SIMILARITY MEASURE (2) Query Record#3 0.96 Record#1 0.80 more similar Record#4 0.77 Record#2 0.63
    • CHALLENGE•  What is “similar”? depends&on&users/tasks& Record#1 (Query) A B C Record#2 missing A B C Record#3 extra A B C D Record#4 Time difference A B C time
    • MATCH & MISMATCH MEASURE time Record#1 (Query) A C B C Record#2 A B C B Matched Events Missing Event Extra Event } Time Difference Number of Swapping Total Score Number of Missing Events 0 to 1 Number of Extra Events
    • RELATED WORK•  Similarity measure –  For numerical time series •  [Ding et al. 2008], [Liu et al. 2006], [Berndt and Clifford 1994], ... –  Edit distance •  [Levenshtein 1966], [Winkler 1999], [Chen 2004], ... –  Biological sequence searching •  [Pearson and Lipman 1988], [Altschul et al. 1990], ... –  Approximate graph matching •  [Tian et al. 2007], [Tian et al. 2008] , ... –  Temporal Categorical Data •  [Sherkat 2006], ...
    • PROTOTYPE: SIMILAN
    • DEMO - DATA•  Patient transfers in the emergency department•  Real data from hospital (Jan – Mar 2010) ADMIT Admission time 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
    • DEMO - SIMILANThis is when bugs are more likely to occur.!
    • CONTROLLED EXPERIMENT•  18 participants•  2 interfaces by 5 tasks –  Similarity-based Search (Similan) –  Exact Search (LifeLines2)•  Time, error, preference
    • X = Exact Search , S = Similarity-based SearchRESULTS Sequence Counting Time const. Uncertainty Sequence CountSpeed (seconds) Time Uncertain60 15 150 15040 10 100 10020 5 50 50 0 0 0 0 X S X S R X S X S Preference (7-points likert scale)8 8 8 86 6 6 64 4 4 42 2 2 20 0 0 0 X S X S X S R X S
    • FLEXIBLE TEMPORAL SEARCH(FTS)
    • PROPOSED WORK: FTS•  Draw an example to specify query•  Specify flexible and inflexible constraints MUST HAVE RANGE ? MUST NOT HAVE optional•  Algorithms
    • PROPOSED WORK: FTS (2) •  Combine exact and similarity-based search Primary Binsimilar Pass all constraints Secondary Binsimilar The rest
    • RESEARCH QUESTIONS1.  How to support the users when they are uncertain about what they are looking for?2.  How to provide an overview of event sequences for temporal categorical data?
    • WHEN YOU READ A PAPER…•  What is it about? INFORMATION VISUALIZATION MANTRA OVERVIEW FIRST, ZOOM AND FILTER, THEN DETAILS ON DEMAND
    • RELATED WORK•  Single-record visualizations Patient ID: 45851737 12/02/2008&14:26 &Admit& 12/02/2008&14:26 &Emergency& Single-record" 12/02/2008&22:44 &ICU& Visualization 12/05/2008&05:07 &Floor& 12/08/2008&10:02 &Floor& 12/14/2008&06:19 &Exit& &•  E.g. LifeLines, MIDGAARD, etc.
    • RELATED WORK (2)•  Multiple instances of single-record visualizations Patient ID: 45851737 Patient ID: 45851737 Patient ID: 45851737 12/02/2008&14:26 &Admit& 12/02/2008&14:26 45851737 Patient ID: &Admit& 12/02/2008&14:26 &Emergency& 12/02/2008&14:26 &Admit& 12/02/2008&22:44 &Emergency& 12/02/2008&14:26 &ICU& 12/02/2008&14:26 &Admit& 12/02/2008&14:26 &Emergency& 12/02/2008&22:44 &ICU& 12/02/2008&14:26 &Emergency& Visualization 12/05/2008&05:07 &Floor& 12/02/2008&22:44 &ICU& Visualization Single-record" Visualization 12/05/2008&05:07 &Floor& 12/02/2008&22:44 &ICU& 12/08/2008&10:02 &Floor& 12/05/2008&05:07 &Floor& Visualization 12/08/2008&10:02 &Floor& 12/05/2008&05:07 &Floor& 12/14/2008&06:19 &Exit& 12/08/2008&10:02 &Floor& & 12/14/2008&06:19 &Exit& 12/08/2008&10:02 &Floor& 12/14/2008&06:19 &Exit& & 12/14/2008&06:19 &Exit& & &•  E.g. LifeLines2, Continuum, ActiviTree, etc.
    • More space please....
    • RELATED WORK (3)•  Multiple-record visualizations Patient ID: 45851737 Patient ID: 45851737 Patient ID: 45851737 12/02/2008&14:26 &Admit& 12/02/2008&14:26 45851737 Patient ID: &Admit& 12/02/2008&14:26 &Emergency& 12/02/2008&14:26 &Admit& 12/02/2008&22:44 &Emergency& 12/02/2008&14:26 &ICU& 12/02/2008&14:26 &Admit& 12/02/2008&14:26 &Emergency& Multiple-record" 12/02/2008&22:44 &ICU& 12/02/2008&14:26 &Emergency& 12/05/2008&05:07 &Floor& 12/02/2008&22:44 &ICU& Visualization 12/05/2008&05:07 &Floor& 12/02/2008&22:44 &ICU& 12/08/2008&10:02 &Floor& 12/05/2008&05:07 &Floor& 12/08/2008&10:02 &Floor& 12/05/2008&05:07 &Floor& 12/14/2008&06:19 &Exit& 12/08/2008&10:02 &Floor& & 12/14/2008&06:19 &Exit& 12/08/2008&10:02 &Floor& 12/14/2008&06:19 &Exit& & 12/14/2008&06:19 &Exit& & &•  E.g. LifeLines2, Continuum
    • LifeLines2’s Temporal Summary [Wang et al. 2009] Continuum’s Histogram [Andre 2007] Not so useful with many event typesDoes not help exploring event sequences
    • OVERVIEW OF EVENT SEQUENCES•  Scalability vs. Loss of information Patient ID: 45851737 Patient ID: 45851737 Patient ID: 45851737 12/02/2008&14:26 &Admit& 12/02/2008&14:26 &Admit& 12/02/2008&14:26 &Emergency& ? 12/02/2008&14:26 &Admit& 12/02/2008&22:44 &Emergency& 12/02/2008&14:26 &ICU& 12/02/2008&14:26 &Emergency& 12/02/2008&22:44 &ICU& 12/05/2008&05:07 &Floor& 12/02/2008&22:44 &ICU& 12/08/2008&10:02 &Floor& 12/05/2008&05:07 &Floor& 12/05/2008&05:07 &Floor& 12/08/2008&10:02 &Floor& 12/14/2008&06:19 &Exit& 12/08/2008&10:02 &Floor& 12/14/2008&06:19 &Exit& & 12/14/2008&06:19 &Exit& & &•  Goal: Show common trends, outliers
    • LIFEFLOW AGGREGATEMerge multiple records into tree VISUALIZE Display the tree
    • AGGREGATE•  Aggregate by prefix #1 #2 #3 #4 Example with 4 records
    • AGGREGATE•  Aggregate by prefix #1 #2 #3 #4
    • VISUALIZE•  Inspired by the Icicle tree Number of files!
    • VISUALIZE (2)•  Modify: Use horizontal axis to represent time•  Video
    • DEMO - LIFEFLOWWhen the lines are combined into flow!
    • LIFEFLOW•  Overview –  See trends –  Spot outliers•  Query•  Compare data
    • COMPARISON ICU ICU Intermediate Intermediate Jan-Mar 2008 April-June 2008Floor
    • PROPOSED WORK: LIFEFLOW•  Designing interaction
    • PROPOSED WORK: LIFEFLOW•  Primary –  Including non-temporal attributes •  Age, Gender, etc. –  Link with Flexible Temporal Search•  Optional –  Rank-by-feature –  Complex aggregation –  Multiple alignments
    • PROPOSED WORK: LIFEFLOW•  Including non-temporal attributes –  [Pinto et al. 2001] Male Male Female Male Male Female
    • PROPOSED WORK: LIFEFLOW•  Complex aggregation #7 #8 Not just merge by prefix #9 #10 Do not merge if the time gap is too different
    • PROPOSED WORK: LIFEFLOW•  Multiple alignments ICU Floor ICU
    • EVALUATION PLAN•  Flexible Temporal Search –  Controlled Experiment •  18-20 participants / 60-90 mins•  LifeFlow –  Usability Study •  15-20 participants / 60-90 mins –  MILCs •  3-5 domain experts / 5-12 weeks
    • CONCLUSIONSummary and expected contributions!
    • EXPLORATORY SEARCH MODELUncertain about what theyare looking forKnow exactly what they want Insight, Hypotheses * search Query Data refine, formulate How to display the data? Difficult to understand the event sequences in the data quickly
    • RESEARCH QUESTIONS1.  How to support the users when they are uncertain about what they are looking for? Flexible Temporal Search2.  How to provide an overview of event sequences for temporal categorical data? LifeFlow: an overview visualization
    • EXPECTED CONTRIBUTIONS1.  Design of visual representations, user interfaces and interaction techniques2.  Algorithms for flexible temporal search3.  Evaluation results4.  Open new directions for exploring temporal categorical data
    • ACKNOWLEDGEMENT DR. PHUONG HO, DR. MARK SMITH, DAVID ROSEMAN WASHINGTON HOSPITAL CENTER http://www.whcenter.org NATIONAL INSTITUTES OF HEALTH (NIH) http://www.nih.gov MICHAEL PACK, MICHAEL VANDANIKERCENTER FOR ADVANCED TRANPORTATION TECHNOLOGY LAB (CATT LAB) http://www.cattlab.umd.edu
    • Q&AQuestions?!
    • THANK YOUThank you!