Geospatial Information and
Knowledge Discovery (GIKD) of the
co-PIs and their students:
With
large, diverse geospatial datasets and an integrated, user-centric interface to
analyze user-defined algorithms, the Instrument will give researchers the
opportunity to experiment with spatially-aware information and knowledge
discovery algorithms. This will allow geospatial pattern mining and
discovery of interesting geospatial patterns on spatial objects in rare event detection,
spatio-temporal change and trend detection, and
correlation mining. For example, in rare event detection, unpredictable events are
extremely difficult to detect because they don't occur often or they occur at a
time/location where they are not expected (e.g., detecting traffic congesting
during a disaster situation).
The
Instrument will provide researchers with historical data that
is used to establish a
baseline for dynamic event behavior models,
and scenarios when deviations from the normal model are
identified.
Algorithms can then be built to better enable prediction models. These can be
input into the Instrument for testing and future rare event detection and
prediction. The Instrument will
also aid in the investigation of bootstrapping
techniques [CBO+02, JDC87] and cost-sensitive learning
approaches [Elk01,ZE01] for rare event detection in
spatial data with semantic awareness.
The Disaster and Failure Studies Program at NIST
(the National Institute of Standards and Technology) [http://www.nist.gov/el/disasterstudies/index.cfm]
(see letter) will utilize the Instrument for their Disaster and Failure
Events Data Repository. The Instrument's mapping and query systems will
also enable research in earthquake analytics through enhanced spatio-temporal
querying of earthquake data mashed up with multi-temporal imagery and many
archival and real-time datasets, ranging from demographics to historical
traffic speeds in street segments.
The Mobile Information Systems Research Center at
the University of Illinois at Chicago [http://www.cs.uic.edu/~wolfson/]
(see letter) will utilize the Instrument for their research on moving objects.
The Instrument will
support the FIU effort under the just announced $11.4M USDOT TIGER award (http://cake.fiu.edu/TIGER2013) led by these co-PIs to develop a sustainable community of UniversityCity and to alleviate traffic
congestion. The proposed data collection instrument will further
facilitate the expansion of the TIGER benefits from the UniversityCity
to all of South Florida and make it an adaptable role model for the
Nation.
The utilization of the
Instrument in application leading to Florida road decongestion is referenced in
the letters from the executive leadership of Florida Department of
Transportation Secretary, South Florida Regional Planning Council, Miami-Dade
County, Miami-Dade Expressway Authority, and City of South Miami. The
Instrument will also enable advanced research by our faculty in conjunction
with the DOT Project.
The analysis of the data
managed by the system could result in predictive traffic estimates that would
strengthen the Informed Traveler Program and Applications (ITPA). By analyzing
real-time streams of video of critical traffic intersections, the instrument
will enable ITPA users to be better informed about transportation modes and
routes to best meet their needs.
A particular research
direction is multi-modal traffic modeling: Car, bus, bicycle and
pedestrian traffic congest subject area during peak hours and cause significant
loss of time and energy. The congestion is due to both the intensity of traffic
and the interference between the different flows of traffic (pedestrian, cars,
etc.).
We will investigate the alleviation of this
problem in two steps using the MRI instrument. The first step is to develop a
simulation of traffic using the MRI Instrument's advanced aerial-image
processing. Advanced aerial imaging is a non-invasive and inexpensive
traffic-sensory method which can be much more efficient than ground-level
cameras or car detectors. The MRI Instrument will store and perform feature
identification on O(10 minute)-frequency
high-resolution balloon aerial images.
We will utilize the feature-recognition of the
Instrument to identify cars, pedestrians, bicycles and buses. These features
will be abstracted from the images by the Instrument and be used to develop a
transit simulation model for the area. The traffic model/simulation will then
be used in the ITPA. The system would
consist of management of traffic-lights and recommendation of parking lots to
commuters (through their smart-phones) based on real-time analysis of traffic using
the Instrument. The ITPA would act to decongest all forms of traffic
simultaneously by optimizing traffic-light periods within the area and by
making smart recommendations for parking lots along least-congested routes.
References Cited
[CBO+02] N. V. Chawla, K. W. Bowyer, L. O. Hall,
and W. P. Kegelmeyer. "SMOTE: Synthetic
Minority Over-sampling TEchnique." Journal of
Artificial Intelligence Research, 16:321-357, 2002.
[JDC87] A. K. Jain, R. C. Dubes, and C.-C.
Chen. "Bootstrap techniques for error estimation." IEEE Transactions on Pattern
Analysis and Machine Intelligence, 9:628-633, 1987.
[Elk01] C.
Elkan. "The foundations of
cost-sensitive learning." In IJCAI, pages 973-978, 2001.
[ZE01] B.
Zadrozny and C. Elkan. "Learning and making decisions when costs and probabilities are
both unknown. In Proceedings of the seventh ACM SIGKDD international
conference on Knowledge discovery and data mining, pages 204-213, 2001.