Transcript Dimensions - Spatial Database Group

Mapcube
Shashi Shekhar
Computer Science Department, AHPCRC
University of Minnesota
[email protected]
(612) 624-8307
http://www.cs.umn.edu/~shekhar
http://www.cs.umn.edu/research/shashi-group/
Biography Highlights










7/01-now : Professor, Dept. of CS, U. of MN
12/89-6/01 : Asst./Asso. Prof. of CS, U of MN
Ph.D. (CS), M.B.A., U of California, Berkeley (1989)
Member: CTS(since 1990),Army Center, CURA
Author: “A Tour of Spatial Database” (Prentice Hall,
2002) and 100+ papers in Journals, Conferences
Editor: Geo-Information(2002-onwards), IEEE
Transactions on Knowledge and Data Eng.(96-00)
Program chair: ACM Intl Conf. on GIS (1996)
Tech. Advisor: UNDP(1997-98), ESRI(1995), MNDOT
GuideStar(1993-95 on Genesis Travlink)
Grants: FHWA, MNDOT, NASA, ARMY, NSF, ...
Supervised 7+ Ph.D Thesis (placed at Oracle, IBM TJ
Watson Research Center etc.), 30+ MS. Thesis
Research Interests





Knowledge and Data Engineering
Spatial Database Management
Spatial Data Mining(SDM) and Visualization
Geographic Information System
Application Domains : Transportation,
Climatology, Defence Computations
Spatial Data Mining, SDBMS

Historical Examples



London Cholera (1854)
Dental health in Colorado
Current Examples





Environmental justice
Crime mapping - hot spots (NIJ)
Cancer clusters (CDC)
Habitat location prediction (Ecology)
Site selection, assest tracking, spatial outliers
Project:
Traffic Database System



Sponsor and time-period: MNDOT, 1998-1999
Students: Xinhong Tan, Anuradha Thota
Contributions to Transportation Domain
Reduce response of queries from hours to minutes
Performance tuning (table design, index selection)


Contributions to Computer Science
GUI design for extracting relevant summaries
 Evaluate technologies with large dataset

Map of Station in Mpls
Gui Design

http://www.cs.umn.edu/research/shashi-group/TMC/html/gui.html
Existing Table
Fivemin
Detector
ReadDate
Time
Dayofweek
Volume
Occupancy
Validity
Speed
Benchmark Queries
1. Get 5-min Volume, occupancy for detector ID = 10 on
Oct. 1st, 1997 from 7am to 8am
2. Get 5-min volume, Occupancy for detector ‘5’ on Aug1
1997.
3. Get 5-min volume, Occupancy for detector ‘5’ on Aug1
1997 from 6.30am to 7.30am.
4. Get average 5-min volume, occupancy, for Monday in
Aug1997 between 8.00 - 8.05,8.05-8.10 …… 9.00
5. Get maximum volume, Occupancy for detector ‘5’ on
Aug1 1997 from 6am to 7am
6. Get the average of AM rushhour hourly volume for a set
of stations on highway I35W-NB with milepoint between
0.0 and 4.0 from Oct. 1st, 1997 to Oct. 5th , 1997
Conclusion
Examples of the Query

Example1:

Query description:


Get 5-min Volume, occupancy for detector ID = 10 on Oct. 1st,
1997 from 7am to 8am
SQL statement:






SELECT readdate, time, xtan.fivemin.detector, occupancy, volume
FROM xtan.fivemin, xtan.datetime
WHERE ReadDate = to_date('01-OCT-97', 'DD-MON-YYYY')
AND time BETWEEN '0705' AND '0800'
AND xtan.fivemin.Detector = '10'
AND xtan.fivemin.
Examples of the Query

Query result 1:
Examples of the Query

Example2:

Query description:


Get the average of AM rushhour hourly volume for a set of stations
on highway I35W-NB with milepoint between 0.0 and 4.0 from Oct.
1st, 1997 to Oct. 5th , 1997
SQL statement:
 SELECT hour, xtan.v_stat_hour.station, avg(volume)
 FROM tan.v_stat_hour, xtan.statrdwy
 WHERE ReadDate BETWEEN to_date('01-OCT-97','DD-MONYYYY') AND to_date('05-OCT-97','DD-MON-YYYY')
 AND hour BETWEEN '06' AND '09'
 AND statrdwy.route = 'I35W-I'
 AND statrdwy.mp >= 0.0
 AND statrdwy.mp <= 4.0
 AND xtan.v_stat_hour.station = statrdwy.station
 GROUP BY xtan.v_stat_hour.station, hour
Examples of the Query

Query result 2:
Project:
Traffic Data Visualization



Sponsor and time-period: USDOT/ITS Inst., 2000-2001
Students: Alan Liu, CT Lu
Contributions to Transportation Domain



Allow intuitive browsing of loop detector data
Highlight patterns in data for further study
Contributions to Computer Science


Mapcube - Organize visualization using a dimension lattice
Visual data mining, e.g. for clustering
Motivation for Traffic Visualization

Transportation Manager



Traffic Engineering





Where are the congestion (in time and space)?
Which of these recurrent congestion?
Which loop detection are not working properly?
How congestion start and spread?
Traveler, Commuter




How the freeway system performed yesterday?
Which locations are worst performers?
What is the travel time on a route?
Will I make to destination in time for a meeting?
Where are the incident and events?
Planner and Research


How much can information technique to reduce congestion?
What is an appropriate ramp meter strategy given specific evolution of
congestion phenomenon?
Dimensions

Available
•
•
•
•

TTD : Time of Day
TDW : Day of Week
TMY : Month of Year
S : Station, Highway, All Stations
Others
• Scale, Weather, Seasons, Event types, …
Comparison with IWEDA

Summary of IWEDA Weather Visualizations


Dimension = system of components, time slot, space
User chooses a system component and a timeslot





It is querying a time slice
Possibilities with mapcube

Other visualizations are facilitated




A cell in the matrix of systems x timeslots
Select a component from the system
User gets a weather map
Changes in weather for a day for a location
Changes in weather for a day for a given route
…
Possibilities with Spatial Data Mining



Co-location of micro phenomena with terrain types
Spatial outliers or discontinuities
Hotspots, e.g., tornado alley
Mapcube :
Which Subset of Dimensions ?
TTDTDWTMYS
TTDTDWS
TTDTDW
TDW
S
STTD
TTD
TDW
S
Next Project
Data Fusion levels and Mapcube


Different Sub-cubes help with different data fusion levels
Level 0: Single Sensor


Level 1: Correlating Multiple Sensors



Local weather as a function of time
Map of spatial variation in weather
Space-time plot for a route for a day
Level 2: Interpret, Aggregate



Detect spatial discontinuities, spatial outliers
Group sensors with similar weather measurements
Group timeslots with similar weather measurements
Singleton Subset : TTD
Configuration:  X-axis: time of day; Y-axis: Volume
 For station sid 138, sid 139, sid 140, on 1/12/1997
Trends:
 Station sid 139: rush hour all day long
 Station sid 139 is an S-outlier
Singleton Subset: TDW

Configuration:

X axis: Day of week; Y axis: Avg. volume.
For stations 4, 8, 577
Avg. volume for Jan 1997

Trends:
Friday is the busiest day of week
Tuesday is the second busiest day of week
Singleton Subset: S
Configuration:  X-axis: I-35W South; Y-axis: Avg. traffic volume
 Avg. traffic volume for January 1997
Trends?:

High avg. traffic volume from Franklin Ave to Nicollet Ave

Two outliers: 35W/26S(sid 576) and 35W/TH55S(sid 585)
Dimension Pair: TTD-TDW
Configuration:
Trends:



X-axis: time of date; Y-axis: day of Week

f(x,y): Avg. volume over all stations for Jan 1997, except
Jan 1, 1997

Evening rush hour broader than morning rush hour
Rush hour starts early on Friday.
Wednesday - narrower evening rush hour
Dimension Pair: S-TTD
Configuration:



X-axis: Time of Day
Y-axis: Route
f(x,y): Avg. volume over all stations for
1/15, 1997
Trends:



3-Cluster
• North section:Evening rush hour
• Downtown area: All day rush hour
• South section:Morning rush hour
S-Outliers
• station ranked 9th
• Time: 2:35pm
Missing Data
Dimension Pair: TDW-S
Configuration:
Trends:




X-axis: stations; Y-axis: day of week

f(x,y): Avg. volume over all stations for Jan-Mar 1997
Busiest segment of I-35 SW is b/w Downtown MPLS & I-62
Saturday has more traffic than Sunday
Outliers – Route branch
Post Processing of cluster patterns
Clustering Based Classification:
 Class 1: Stations with Morning Rush Hour
 Class 2: Stations Evening Rush Hour
 Class 3: Stations with Morning + Evening Rush Hour
Triplet: TTDTDWS: Compare Traffic Videos
Configuration: Traffic volume on Jan 9 (Th) and 10 (F), 1997
Trends:
 Evening rush hour starts earlier on Friday
 Congested segments: I-35W (downtown Mpls – I-62);
I-94 (Mpls – St. Paul); I-494 ( intersection I-35W)
Size 4 Subset: TTDTDWTMYS(Album)
Configuration:
Trends:
 Outer: X-axis (month of year); Y-axis (route)
 Inner: X-axis (time of day); Y-axis (day of week)
 Morning rush hour: I-94 East longer than I-35 W North
 Evening rush hour: I-35W North longer than I-94 East
 Evening rush hour on I-94 East: Jan longer than Feb