Information Technology Reference
In-Depth Information
results are still valid for larger processes where individual segments undergo
evolution one at a time. In that case our results describe the behavior for a
single segment.
7Con lu on
Users in collaboration and coordination intensive people-driven processes re-
quire self-learning mechanisms to reflect the evolution of message types in the
underlying process model. In this paper we presented an approach based on a
probabilistic process model and message state model to predict when and which
messages will arrive to ultimately give suitable activity recommendations. Anal-
ysis of sequence logs containing both messages and activities allows updating the
process model automatically. Evaluation based on a motivating scenario demon-
strated successfully that our mechanisms work correctly and eciently.
Future work consists of two main tasks: on the one hand we aim at improv-
ing the message prediction algorithm by extracting message patterns such as
request-reply from interaction logs, including temporal aspects, and integrating
the process learning techniques introduced in our previous work. Pairing the
probabilistic algorithms with semantic message analysis is expected to further
improve the classification success rate. On the other hand, we plan to conduct
additional user studies based on a larger process and log sequences set.
Acknowledgment.
This work has been partially supported by the EU STREP
project Commius (FP7-213876) and Austrian Science Fund (FWF) J3068-N23.
References
1. Adams, M., Edmond, D., ter Hofstede, A.H.M.: The application of activity the-
ory to dynamic workflow adaptation issues. In: 7th Pacific Asia Conference on
Information Systems, pp. 1836-1852 (2003)
2. Adams, M., Hofstede, A., Edmond, D., van der Aalst, W.: Facilitating flexibility
and dynamic exception handling in workflows through worklets. In: Proceedings of
the CAiSE 2005 Forum, FEUP, pp. 45-50 (2005)
3. Burkhart, T., Loos, P.: Flexible business processes - evaluation of current
approaches.
Proceedings
Multikonferenz
Wirtschaftsinformatik,
MKWI-2010
(February 2010)
4. Burkhart, T., Werth, D., Loos, P.: Commius - An Email Based Interoperability
Solution Tailored For SMEs. Journal Of Digital Information Management 6 (2008)
5. Dorn, C., Burkhart, T., Werth, D., Dustdar, S.: Self-adjusting recommendations
for people-driven ad-hoc processes. In: Proceedings of International Conference on
Business Process Modelling. Springer, Heidelberg (September 2010)
6. Dustdar, S.: Caramba Process-Aware Collaboration System Supporting Ad hoc and
Collaborative Processes in Virtual Teams. Distributed Parallel Databases 15(1),
45-66 (2004)
7. Greenwood, D., Rimassa, G.: Autonomic goal-oriented business process manage-
ment. In: ICAS 2007: Proceedings of the Third International Conference on Au-
tonomic and Autonomous Systems, p. 43. IEEE Computer Society, Washington,
DC, USA (2007)
