time-series data kaitlin duck sherwood cs 533c. why do you care? time-series data is all over the...
TRANSCRIPT
![Page 1: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/1.jpg)
Time-Series Data
Kaitlin Duck Sherwood
CS 533c
![Page 2: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/2.jpg)
Why do you care?
• Time-series data is all over the place.
![Page 3: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/3.jpg)
What is Time-Series Data?
• Lines.
![Page 4: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/4.jpg)
What is Time-Series Data?
• Usually periodic
Source: van Wijk and van Selow, Cluster and Calendar based Visualization of Time Series Data, 1999
![Page 5: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/5.jpg)
Spiral Viewer (Carlis et al)
• Angle position in cycle• Radius cycle number
Color, diameter available for use
![Page 6: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/6.jpg)
Unknown periodicity
• Tweak period in realtime to find periodicity
• Example: music
![Page 7: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/7.jpg)
Helices
• Example: Chimp eating habits
• Angle day of year
• Radius year
• Z-axis* type of food
• Color type of food
• Diameter amount
(Rings to beat occlusion)
![Page 8: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/8.jpg)
Spiral barchart
Z-axis used for data
Can step through spokes one at a time
![Page 9: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/9.jpg)
User feedback
• Qualitative feedback from 12 users
• “Buy into the notion of a spiral display”
• Couldn’t self-operate
• Wanted more
![Page 10: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/10.jpg)
Good points of paper
• Compelling visuals
• Gave examples
• Has software
• Some user feedback
![Page 11: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/11.jpg)
Bad points of paper
• Examples not compelling
• Graphs unlabeled
• Difficult to see quant info
• Questionable movie data
• Weak user eval
• Advantages over Cartesian?
![Page 12: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/12.jpg)
Time-series bitmaps
Repurposes heavily:
• Chaos Game Representation
• SAX
• Windows Explorer
![Page 13: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/13.jpg)
Chaos Game Representation• Assign corner of square to each base
• For each symbol, take a step in symbol direction of half distance
• Color corresponds to number of times a pixel visited
![Page 14: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/14.jpg)
SAX
• Converts reals into equiprobable letters
• Eliminate trends with narrow window
• Uses: clusters, motifs, anomalies
![Page 15: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/15.jpg)
Time-series bitmaps
• Data -> SAX -> CGR -> bitmap
• Linear color mapping (JET)
• Length normalization
![Page 16: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/16.jpg)
Ubiquity
Use filesystem:• Thumbnails• Cluster (using
MDS)
Comparisons only
![Page 17: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/17.jpg)
Real-world data
• Clustering heterogenous data (15 sets)– Better than ARIMA or Markov
• Clustering of 20 ECG patients (perfect)
• Video classification – Better than Euclidian or DTW
• Classifying ECG data (perfect)
![Page 18: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/18.jpg)
Good points
• Pretty cool idea
• Repurposes material from other fields well
• Ubiquitous visualization (filesys)
• Impressive results
![Page 19: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/19.jpg)
Bad points
• Didn’t explain CGR well
• Didn’t explain Windows clustering
• Data sets relatively small
• No user testing
![Page 20: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/20.jpg)
Summary
• Spirals. Cool pictures, what use?
• Bitmaps. Less cool, perhaps more useful.
![Page 21: Time-Series Data Kaitlin Duck Sherwood CS 533c. Why do you care? Time-series data is all over the place](https://reader036.vdocuments.mx/reader036/viewer/2022070323/56649d995503460f94a8414a/html5/thumbnails/21.jpg)
References
• Interactive Visualization of Serial Periodic Data, John V. Carlis and Joseph A. Konstan, Proc UIST 98.
• Time-series Bitmaps: A Practical Visualization Tool for working with Large Time Series Databases Kumar, N., Lolla N., Keogh, E., Lonardi, S. , Ratanamahatana, C. A. and Wei, L. (2005). Proc. SDM '05, pp. 531-535