Expert systems and artificial intelligence technologies have a significant
role to play in the development of tools to support police operations. Syst
ems have been successfully demonstrated in the areas of breaking and enteri
ng, criminal profiling, tracking serial criminals and others. In this paper
, we describe the development of a system that will support police investig
ations in several crime areas. The methodology has been adapted from that u
sed in several North American breaking and entering projects, to meet the r
equirements of the Hong Kong police. The system is initially rule-based, wi
th provisions for investigating machine-learning and neural network techniq
ues to improve performance and makes extensive use of existing data sources
. The development of the initial demonstration system is described and the
methodology is discussed. (C) 1998 Elsevier Science B.V. All rights reserve
d.