Top Banner
Interesting Interval Discovery on Spatiotemporal Datasets Csci 8715 Fall 2013
5

Interesting Interval Discovery on Spatiotemporal Datasets Csci 8715 Fall 2013.

Dec 28, 2015

Download

Documents

Jason Lang
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Interesting Interval Discovery on Spatiotemporal Datasets Csci 8715 Fall 2013.

Interesting Interval Discovery on Spatiotemporal Datasets

Csci 8715 Fall 2013

Page 2: Interesting Interval Discovery on Spatiotemporal Datasets Csci 8715 Fall 2013.

Sub-path of Abrupt Change

• Spatial sub-path of Abrupt Change[1]– Sharp change in vegetation cover– Transition between ecological zones (ecotones)– Vulnerable to climate change– Moves in response to climate change

The change is persistent and rapid

W1=[12N, 17N]

A snapshot of vegetation cover in Africa [6]

Page 3: Interesting Interval Discovery on Spatiotemporal Datasets Csci 8715 Fall 2013.

Sub-path of Abrupt Change

• Temporal sub-path (interval) of Abrupt Change[1]৹ Abrupt shift in precipitation, temperature, etc. ৹ Climate change detection.

Raw Sahel precipitation anomaly (JJASO) Smoothed Sahel precipitation anomaly (JJASO)

Page 4: Interesting Interval Discovery on Spatiotemporal Datasets Csci 8715 Fall 2013.

Case Study Output

AVG{∆}AVG≥α{∆}

• Data: NDVI by GIMMS, Africa, 1981 August. Resolution: 8km. Smoothed within 1x1 degree.• Path: along each longitude (south north)• Interest measure: (Slope) Sameness degree

– ∆ : unit slope

• Thresholds: α= 20% percentile, SD ≥0.5

Page 5: Interesting Interval Discovery on Spatiotemporal Datasets Csci 8715 Fall 2013.

Potential Project ideas

• Explore other interest measures to find interesting interval patterns– High correlations time periods between two time series– High variation periods of time series

• Implement an R software[2] package and enable user defined interest measure

[1]. Xun Zhou, Shashi Shekhar, Pradeep Mohan, Stefan Liess, and Peter K. Snyder. "Discovering interesting sub-paths in spatiotemporal datasets: A summary of results." In Proceedings of the 19th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, pp. 44-53. ACM, 2011.

[2]. The R project for statistical computing: http://www.r-project.org/