出版社：Academy & Industry Research Collaboration Center (AIRCC)
摘要：Regulatory processes are normally tracked by regulatory bodies in terms of monitoring safety,
soundness, risk, policy and compliance. Such processes are loosely framed processes and it is a
considerable challenge for data scientists and academics to extract instances of such processes
from event records and analyse their characteristics e.g. if they satisfy certain process
compliance requirements. Existing approaches are inadequate in dealing with the challenges as
they demand both technical knowledge and domain expertise from the users. In addition, the
level of abstraction provided does not extend to the concepts required by a typical data scientist
or a business analyst. This paper extends a software framework which is based on a semantic
data model that helps in deriving and analysing regulatory reporting processes from event
repositories for complex scenarios. The key idea is in using complex business-like templates for
expressing commonly used constraints associated with the definition of regulatory reporting
processes and mapping these templates with those provided by an existing process definition
language. The efficiency of the architecture in evaluation, compliance and impact was done by
implementing a prototype using complex templates of Declare ConDec language and applying it
to a case study related to process instances of Australian Company Announcements.
关键词：Regulatory Reporting; Process Extraction; Semantic Technology; Events.
