General Systems and Theory Papers
A Relation Context Oriented Approach to Identify Strong Ties in Social Networks
HIT: A GIS-Based Hotspot Identification Taxonomy
Using an Edge-dual Graph and k-connectivity to Identify Strong Connections in Social Networks
Utilizing Commodity Hardware and Software to Distribute a Real-World Application: Maximizing Reuse While Improving Performance
Optimizing Disk Storage to Support Statistical Analysis Operations
Strategies to Improve Variable Selection Performance
*****
A Relation Context Oriented Approach to Identify Strong Ties in Social Networks
Li Ding, Dana Steil, Brandon Dixon, Allen Parrish, David Brown;
Annals of Information Systems, Oct. 2009
Social network graphs have been found to be an extremely effective tool in
the identification of potential perpetrators of criminal activity. These
graphs can grow extremely large, as illustrated by an example within this paper
that contained over 4.9 million nodes and over 211 million edges.
Obviously some reduction of these graphs is essential to their being useful.
Further, considerable “noise” (false positive relationships) are generated when
the graphs are totally comprehensive. This research transformed the
original social network into a relational context-oriented edge-dual graph.
This was done by evaluating the quality of the connectivity for each edge to
obtain a metric to this effect for each edge. By retaining only the
strongest edges the overall graph becomes more reliable and more useful in
practice.
For the complete paper click here.
HIT: A GIS-Based Hotspot Identification Taxonomy
Steil, D. and A. Parrish;
International Journal of Computers and Their Applications, to appear, 2009.
The authors have developed a Hotspot Identification Taxonomy (HIT)
that organizes the various methods for viewing hotspots. Basically
they are defined as follow:
- First order - high crash frequency road segments possibly filtered for specific event(s);
- Second order - road segments defined as those that have high event counts specifically related to a countermeasure under consideration (e.g., selective enforcement for the speed event);
- Third order - segments having a high frequency of countermeasure-related events and for which the countermeasure was historically effective.
Effective use of the HIT model required four interrelated activities:
data-collection, linear hotspot identification, presentation and assessment.
For the complete paper click here.
Using an Edge-dual Graph and k-connectivity to Identify Strong Connections in Social Networks
Li Ding, and Brandon. Dixon,
in Proc. ACM Southeast Regional Conference 2008, Auburn, Alabama, US 2008
The goal of this paper is to use edge-dual graph transformation techniques to
improve the accuracy of social network analysis (SNA). SNA is used in law enforcement
to determine if relationships exist among potential suspects, and to identify just what
those relationships might be. Relationships can be family, friends, past associates, cell
mates and even prison enemies. The paper presented results that showed that this
transformation had a very high potential for increasing the accuracy of relationship search routines.
For the complete paper click here.
Utilizing Commodity Hardware and Software to Distribute a Real-World Application: Maximizing Reuse While Improving Performance
Davis, M., R. Smith, B. Dixon, A. Parrish and D. Cordes,
Software: Practice and Experience, Volume 35, no. 7, June 2005.
This research delved into the current use of the commodity computing hardware,
which is motivated by a dramatic increase in the performance to price ratio. The
research evaluated the performance of a statistical analysis application in a ten-node
off the shelf computing cluster. The study had two stems: (1) examining the various
network topologies, and (2) minimizing the software modifications required in distributing
the application. The general conclusion was that when reuse of existing code is feasible,
performance can be dramatically increased by the combined use of parallel computing and
commodity components.
For the complete paper click here.
Disk Storage to Support Statistical Analysis Operations
Parrish, A., S. Vrbsky, B. Dixon and W. Ni, database techniques generally require the reading of complete rows of data (traditionally
referenced as “records”) in order to get at a single attribute that might be of interest.
Further, if filtering is required (not all records are of interest), a further computational
step is needed on each record to determine if it qualifies. Transposition of the data enables
this to be accomplished with a single read operation, followed by a single filter-pointer
operation producing essentially instantaneous results. This method has proven successful
in producing real-time instantaneous results when applied to well over millions of records.
For the complete paper click here.
Strategies to Improve Variable Selection Performance
Wang, H., A. Parrish, R. Smith, S. Vrbsky,
Proceedings of the 2005 International Conference on Information and Knowledge Engineering, Las Vegas, June 2005
This paper compares a “row major order” data structure that is used standard relational
databases against the transposed “column major order” data structure used by CARE.
These data structures are described in detail, as were the various filtering methods
that could be employed. Performance tradeoffs between the two data structures
demonstrated a clear advantage of the column major order over the traditional storage approaches.
For the complete paper click here.
⇑TOP⇑