Collaborative adaptive case management with linked data

Sebastian Heil, Stefan Wild, Martin Gaedke

Abstract: An increasing share of today's work is knowledge work. Adaptive Case Management (ACM) assists knowledge workers in handling this collaborative, emergent and unpredictable type of work. Finding suitable workers for specific functions still relies on manual assessment and assignment by persons in charge, which does not scale well. In this paper we discuss a tool for ACM to facilitate this expert finding leveraging existing Web technology. We propose a method to automatically recommend a set of eligible workers utilizing linked data, enriched user profile data from distributed social networks and information gathered from case descriptions. This semantic recommendation method detects similarities between case requirements and worker profiles. The algorithm traverses distributed social graphs to retrieve a ranked list of suitable contributors to a case according to adaptable metrics. For this purpose, we introduce a vocabulary to specify case requirements and a vocabulary to describe skill sets and personal attributes of workers. The semantic recommendation method is demonstrated by a prototypical implementation using a WebID-based distributed social network.