There are
several Research projects developed in the Engineering Cell of Knowledge, in
UNIFOR, under the orientation of Dr. Vasco Furtado. Work in carried out on
the aspects linked to the acquisition, representation and exploration of
knowledge, for the production of intelligent systems, focusing applications
linked to the Administration of Knowledge, the Mining of Data in Grid, Web
Semantics and Technologies for Public Security.
Projects
of the Area of Law Enforcement:
One of the crucial questions regarding crime and violence control in urban
centers is how to gauge the actual impact of certain police management
strategies upon the regulation of the rates of criminality. This is indeed a
question difficult to be answered, as it seems that the effectiveness of a
certain law-enforcement policy on a given metropolitan region depends,
directly or indirectly, upon an array of factors, ranging from the levels of
concentration of richness to the physical organization of the urban center
under consideration. In such a context, it is quite consensual that police
patrolling can be considered as one of the best well-known means for
implementing preventive strategies towards the combat of an assortment of
crimes, mainly those involving violence practices (such as bank assaults,
thefts, armed robbery, gang fights, drug dealing, environmental degradation,
and kidnapping).
The conceptual basis for preventive approaches and the development of some
pro-active policing strategies can be found in “Routine Activities
Theory” (Cohen and Felson, 1979), which attempts to explain the
evolution of crime rates not only through the characteristics (psychological
profiles) of the offenders, but also through the circumstances in which crimes
occur. Basically, Cohen and Felson (1979) point out that, in order for a
criminal act to take place, three elements must coexist: a motivated
offender; a suitable target, either an object or person that can be attacked;
and the absence of capable guardians, in charge of the preventive actions.
The criminality model derived by the authors is then based on an economic
equation involving the aforementioned elements. A direct conclusion of such
work is that criminal offenses are related to the nature of everyday patterns
of social interaction. Another is that the police force is, naturally, the
central element for promoting public safety by means of diligence and
dissuasion.
In our work, we have investigated some behavioral aspects of criminality by
making use of agent-based simulation having in mind both educational and
decision-support purposes. Agents representing criminals, police officers,
and targets, as taking part in a multiagent society, had their main
properties, profiles, and behaviors characterized and modeled. We are
investigating different approaches to different problems in police patrol.
They are:
ExpertCOP
Using Genetic Algorithms to Discover Police Patrol Routes
Analysis of Preventive Routes Reorganization
Data Mining - FINEP
Integration of Geo-processing techniques to the tool
SmartBase Viewer
Explanation
of Problem Solving Methods:
Generation User Interfaces
Explanation of Problem Solving Methods
Explanation of composition of Web Services performed by
agents solving problems
Data
Mining in Grid:
A Knowledge-Based Architecture for Helping in the Optimization and
Development of Data Mining Applications in Grids
Use of Aspects for crosscutting concerns in the development
of DM applications for Grid
Agent-mediated
Knowledge Management:
A Process-oriented and Ontology-based Methodology for the
Construction of Multi-Agent Systems to support Knowledge Management