transitioning experiences with army geo spatial center (agc) pradeep mohan 4 th year phd student 1
TRANSCRIPT
Transitioning Experiences with Army Geo Spatial Center (AGC)
Pradeep Mohan
4th Year PhD Student
1
Overview: Best Practices
Practice AGC CrimeStat Libraries 1.0 CrimeStat 3.2
Function Level Documentation
√
Parameter sensitivity analysis √
Pattern Analysis with test datasets
√
Well defined Test Cases √ √
Unit Testing √
Integration Testing √ √
Alpha/ Beta Testing √ √
API Documentation √ √
Testing on Multiple Platforms √
Multiple Data formats √ √ √
2
3
Cascading Spatio-temporal pattern discovery
Stages: Bar Closing, Assault , Drunk Driving, Hurricane, Climate change etc.
Cascading spatio-temporal pattern (CSTP)Bar Closing
Assault
Drunk Driving
Partially ordered subsets of ST event types.
Located together in space.
Occur in stages over time.
Other Applications: Climate change, epidemiology, evacuation planning.
T1 T2 T3
B.2
B.1C.1
C.2 C.3
C.4
A.1
A.3
A.2
A.4
Assault(A)
Drunk Driving (C)
Bar Closing(B)
Aggregate(T1,T2,T3)
C2C.3
C.4
C.1
A.1
A.3
A.2
A.4
B.2B.1
Project: Cascade models for multi-scale pattern discoveryJ.W: Dr. J.A. Shine and Mr. J.P. Rogers (AGC)
[1] Pradeep Mohan, Shashi Shekhar, James A. Shine, James P. Rogers. Cascading spatio-temporal pattern discovery: A summary of results. In Proc. of 10th SIAM International Data Mining (SDM) 2010, Columbus, OH, USA[2] J.A. Shine, J.P.Rogers, S.Shekhar, P.Mohan. Cascade models for multi scale pattern discovery: An Extended abstract. In USARMY ERDC Conference 2009, Memphis, TN
Cascade Pattern Discovery[1]Source Code: Matlab 2009b
Test Case: Crime Data,Parameters
Source code independent of Toolboxes
Toolbox Dependencies
Shape Files
.MAT Files (Test Cases)
AGC
a. Performance analysisb. Pattern analysisc. Parameter sensitivity
a. Patternsb. Performance bottlenecksc. Bugsd. Other issues like visual display
Entire Process~2 Months
4
Project: Cascade models for multi-scale pattern discoveryJ.W: Dr. J.A. Shine and Mr. J.P. Rogers (AGC)
[3] (Ongoing) Pradeep Mohan, Shashi Shekhar, James A. Shine, James P. Rogers. Cascading spatio-temporal pattern discovery: A summary of results. (Journal Version)
AGC Requirements
Pattern Visualization
Performance Enhancement
Fixing Bugs
Parallelizable
Toolbox independence
Our Actions
Pattern Data structure changes
Faster Algorithms[3]
Revised and Tested Code
? MPI Support in Matlab
? Migration to C++
5
CrimeStat
A Spatial Statistics Program for the Analysis of Crime Incident Locations
6
Our Contributions• Crime Stat Libraries 1.0[1]
– Set of .NET components distributed by NIJ
• Crime Stat v 3.2– Statistical Simulation functions for Spatial Analysis Routines
• Scalability to Large Datasets– Self-Join Index[2]
[1] http://www.spatial.cs.umn.edu/projects/crimestat-pub/beta/ [2] Pradeep Mohan, Shashi Shekhar, Ned Levine, Ronald E. Wilson, Betsy George, Mete Celik, Should SDBMS support the join index ?:
A Case Study from Crimestat. In Proc. of 16th ACM SIGSPATIAL International Conference on
Advances in Geographic Information Systems (ACM GIS 2008), California, USA,2008.
7
Project: Crime stat Libraries 1.0J.W: Mr. Ron Wilson (NIJ) and Dr. Ned Levine (Ned Levine and Associates)
Pradeep Mohan~1.5 Years
Entire Process~2 .5Years
Vijay Gandhi~1 Year
Chetan~1 Year
Core ComponentsSpatial Description
Spatial Analysis Spatial Interpolation Journey to Crime Distance Analysis Alpha Testing Feedback
Performance Tuning Outputs in several formats Beta Testing-I Feedback
NIJ+ Beta Testers
Documentation
Alpha TestingFeedback
Algorithm Descriptions Test Cases
Beta Testing-I
Documentation Testing Framework Feedback updatesFeedback
Beta Testing-II
Documentation Visual Outputs Wrap upFeedback
8