=Paper=
{{Paper
|id=Vol-2063/salad-paper3
|storemode=property
|title=Annotating sBPMN Elements with their Likelihood of Occurrence
|pdfUrl=https://ceur-ws.org/Vol-2063/salad-paper3.pdf
|volume=Vol-2063
|authors=Tobias Weller,Maria Maleshkova
|dblpUrl=https://dblp.org/rec/conf/i-semantics/WellerM17
}}
==Annotating sBPMN Elements with their Likelihood of Occurrence==
https://ceur-ws.org/Vol-2063/salad-paper3.pdf
Annotating sBPMN Elements with their
Likelihood of Occurrence
Tobias Weller and Maria Maleshkova
Institute for Applied Informatics and Formal Description Methods, Karlsruhe
Institute of Technology (KIT),
Englerstr. 11, 76133 Karlsruhe, Germany
{tobias.weller,maria.maleshkova}@kit.edu
http://www.aifb.kit.edu
Abstract. Process Mining is a research discipline that aims to analyze
business processes based on event logs. The event logs are among others
used to create models for predicting the next activity of a given process
instance. Existing models use Bayesian Networks or Markov Chains to
predict the next activity in a workflow. These models require knowledge
about the occurence of activities in the business process, which is usually
based on expert knowledge or based on previous workflows from event
logs. Based on previous work, we will i) represent a business process in
sBPMN and extend our annotation tool to ii) compute the likelihood
of occurrence of activities in a business process and check for stochastic
dependency in a process and iii) use the generated knowledge to annotate
the business process.
Keywords: Process Models, sBPMN, Markov Chain, Annotation
1 Introduction
A very common research topic in Process Mining is the prediction of next activ-
ities in a business process. The recommendation of next activities can either be
based on a schema, if a target process has been defined, on statistical methods,
which exploits knowledge from past events, stored in a repository, or as a hybrid
system, both in combination. The persons, involved in the process, get recom-
mendations from a system, which next activities they should perform. Often
are those recommender systems out to optimize a certain optimization function,
which aims at decreasing the mortality of patients, runtime of business process
or increase the satisfaction of involved persons. However, it is often very hard to
predict next activities, because the underlying data is very heterogeneous and
different linguistic explanations for similar activities lead to fault predictions.
In addition, if no target process is defined, then the number of process varia-
tions might be high due to different opinions in the process execution by various
persons. This is among others given in the medical domain. Various physicians
might have different opinions and follow different best practices. This leads to
a high number of process variations for similar processes. A further impact in
predicting next activities is the various number of influences that comes in.
2 Tobias Weller and Maria Maleshkova
Semantics and background knowledge might improve the results of predicting
next activities in a process. So far, including semantics and background knowl-
edge, has not been considered deeply. Often, only the sequence of activities has
been considered. However, there is semantics hidden in the log files of processes
that can be used to improve statistical methods to predict next activities. Often
occurrence of activities depend on each other.
Our previous works focused on finding correlations in meta-information of a
business process. Now we are interested in finding dependencies in the worfklow
of a business process. Our aim is to compute the likelihood of the next activity
and therefore the likelihood of a certain outcome and workflow. We will also check
if events in a business process are stochastically independent. This knowledge
can be used to compute the likelihood of different workflows and predicting the
next activity. By annotating the business process with the likelihoods of their
occurrences, a Markov chain can easily be created [6]. We will use the business
process depicted in figure 1 as consistent example.
Fig. 1. Considered business process for annotating activities with their likelihood of
occurrence.
Based on previous work, we will represent a business process in sBPMN
and i) extend our annotation tool to ii) compute the likelihood of occurrence of
activities in a business process and check for stochastic dependency in a process
and iii) use the generated knowledge to annotate the business process.
In summary, this paper makes the following contributions: 1. Compute the
likelihood of occurrence of activities in a business process and check for stochas-
tically dependencies. 2. Annotate the sBPMN with the generated knowledge.
The remainder of this paper is structured as follows: the following section
(Section 2) introduces the methods used to compute the likelihoods and stochas-
tic dependencies. Section 3 demonstrates the practical applicability of our so-
lution by realizing our approach with open-source process data and evaluating
the added value to recommend next activities and computing the likelihood of
certain workflows. Related work is described in section 4. We sum up our con-
tributions and provide conclusions and future work in section 5.
Annotating sBPMN Elements with their Likelihood of Occurrence 3
2 Material and Methods
Our aim is to annotate activities of a business process with their corresponding
likelihood of occurrence. This knowledge can be used to compute the likeli-
hood of a certain outcome, workflow and use this knowledge in predicting the
next activity. We start from the premise that a target process is already de-
fined in Business Process Model and Notation (BPMN). However, because we
want to exploit semantic relationship and meta-information in future, we are
interested in including the semantic information into BPMN. Semantic Business
Process Model and Notation (sBPMN) allows to add meaning in form of meta-
information and background knowledge to each process elements. The result is
a machine-readable format, which allows for reasoning on the process descrip-
tion. Therefore, we will use sBPMN to model the processes, because it allows to
add semantics to business processes, which is used to describe the information
of the business processes. Hence, a first step is to transform a given business
process in BPMN into sBPMN. For this purpose, we will use previous work that
automatically transforms a BPMN process in the standard format BPMN 2.0
XML by OMG [1] into sBPMN. There are already many ontologies for BPMN
2.0 available [2–5] that allows to capture the semantics in processes and include
meta-information. However, not all ontologies are online available and follow the
latest BPMN 2.0 version. We have to pick a suitable one. By using an ontology,
we can easily add annotations to the BPMN elements.
Critical issues in a process are usually branches in which involved persons
have to decide which way to pick. If no decision criteria is given, then a very
basic approach is to assume a Laplace’s probability space for the next activity.
This means that the likelihood of occurrence of each previous activity is the same
(uniform distribution over the following activities). We exemplary depicted in
figure 2 the likelihoods of the next activity of the branches by using a Laplace’s
probability space assumption. For all other previous activities, without branches,
is the likelihood 1.
However, the assumption of a Laplace’s probability space is for recommending
next activities in a real-world scenario not sufficient. To improve our model, we
will compute the likelihood of occurrence by using existing process instances
from a repository. For this purpose, we use historic data from a repository to
compute the likelihood of the next activities. For sequences of activities, the
next activity is obviously. Therefore, the likelihood of the next activity is 1.
So crucial parts of the process are the branches for which multiple activities
can follow. In our example those are the activities after Check application form
completeness, Assess eligibility, Check if home insurance quote is requested and
Verify repayment agreement.
By using the computed likelihoods, one can easily use them to forecast the
next activity and to model a Markov chain. The assumption of Markov chains
is that the next state depends only on the current state. Which was calculated
above. Therefore, we can use to predict the outcome of a workflow and the like-
lihood of a certain workflow by using the Markov chain assumption. However,
the outcome or occurrence of activities might dependent not only on the current
4 Tobias Weller and Maria Maleshkova
activities or states but on previous activities. E.g. one might think that the oc-
currence of returning the application back to the applicant due to an incomplete
form might lead to a higher likelihood of rejecting the application. In order to
check such dependencies, we will check for stochastically dependencies of activ-
ities in the process. Two events A and B are stochastically independent if the
following applies:
P (A) · P (B) = P (A ∩ B)
3 Evaluation
We used a free available data set from BPI Challenge 2012 [7]. The log files
of each process instance were created synthetically. We used the parallel target
process, which contains 10,000 process instances. Meta-information, except for
time-stamps, are not given. We described the target process by using Cogni-
tive Process Designer [8, 9]. This tool is an extension to Semantic MediaWiki
(SMW) [10] that allows for capturing BPMN diagrams, meta-information about
the activities and describe the information semantically. Semantic MediaWiki is
a powerful collaborative knowledge management system, using the MediaWiki
engine and allowing for capturing information in a structured way. The captured
information are stored by using RDF as standard format and can therefore be
queried. By using an appropriate ontology, the target process is available in
sBPMN and can easily be enriched with further information. We used an ontol-
ogy published by DKM [2]. Afterwards, we uploaded the process instances into
the SMW, in which the Cognitive Process Designer runs. We linked the pro-
cess instances to the activities of the target process. As discussed in section 2,
a Laplace’s probability space might be not applicable in real-world scenarios.
Therefore, we used a bash script, which counts for every occurence of an activ-
ity the likelihood of appearence of the next activity. The fact that the business
process is stored in RDF facilitated this step due to the fact that the number of
activities could easily be queried. We applied this on the data set and received
the results depicted in figure 2. The likelihood for sequences of activities is al-
ways one and not shown on the graphic due to reasons of clarity. Likelihoods of
activites, which are not connected by a direct edge, is zero. Therefore, we only
depicted the likelihoods of the next decision at decision points, because these
are the crucial parts of the process. We calculated the occurrence of an activity
in general as well but not depicted it on the graph.
To comprehend the application, we will show two examples in the following.
In total 1,070 times were the Activity Return application back to applicant chosen
as next activity after Check application form completeness and 10,000 times
Appraise property. This leads to the following likelihoods:
1070
P (Xt = Appraise Property|Xt−1 = Check application) = = 0.0967
11070
10000
P (Xt = Return application|Xt−1 = Check application) = = 0.9033
11070
Annotating sBPMN Elements with their Likelihood of Occurrence 5
Fig. 2. Annotation of workflow by using Laplace approach and expected value for
Markov Chains of next activity on decision points of the considered target process of
the BPI Challenge 2012.
For the next decision point (Asses eligibility), we computed the following likeli-
hoods:
4916
P (Xt = Reject application|Xt−1 = Asses eligibility) = = 0.4916
10000
5084
P (Xt = Prepare acceptance pack|Xt−1 = Asses eligibility) = = 0.5084
10000
The bash script computed stochastically dependency between activities as
well. Therefore, it checks if there might be dependencies between the occurence
of certain activities exist. For instance, one might think of a stochastic depen-
dency if an application were returned due to an incompletness of the form and
a rejection. Therefore we checked the stochastically dependency between these
two events.
963 4916
P (Return application) · P (Reject application) = · = 0.0473
10000 10000
486
P (Return application ∩ Reject application) = = 0.0486
10000
Because 0.0473 ≈ 0.0486, we can assume that the events Return application
and Reject application are stochastically independent. We checked other combi-
nations of events for stochastically dependency as well. E.g. we checked if the
events Return application and Cancel application are stochastically independent.
963 2453
P (Return application) · P (Cancel application) = · = 0.0236
10000 10000
223
P (Return application ∩ Cancel application) = = 0.0223
10000
Due to the fact that 0.0236 ≈ 0.0223, we assume that there is no dependency.
We performed tests by using a sampling set of 9,000 workflows for computing
the likelihoods of the next activities and predicted the next activities by using the
likelihoods. As assumed, because we could not find any dependency in the work-
flows, the likelihoods also reflect the error rate. E.g. we computed the likelihood
6 Tobias Weller and Maria Maleshkova
of the activities, occuring after Check application for completness. The result was
9000
P (Appraise Property|Check application form completness) = = 0.9038,
9958
respectively
P (Return application back to applicant|Check application form completness) =
0.0962. Therefore, as a very basic approach, we always suggested the activity
with the highest likelihood for the evaluation set (1,000 workflows). In this case,
the error rate was 0.1007. Which is the proportion of the likelihood of the total
result ((1000 · 0.1007 + 9000 · 0.0962) ÷ 10000).
We used the generated knowledge about the likelihood of occurence of each
event and the likelihood of the next event in our sBPMN model by including it as
meta-information. We attached these information on the activities and decision
nodes. By using the likelihoods and assuming a Markov chain characteristic,
we can calculate the likelihood of different workflows of the process. E.g. the
likelihood that the the application is returned once to the applicant and then
rejected is 0.0429. This corresponds to the likelihood of this scenario given in
the data (435 times occured this workflow1 ). The respective likelihood that the
application form is returned two times to the applicant and then rejected is
0.0042, which also corresponds to the occurred likelihood in the data (47 times
occured this workflow2 ).
4 Related Work
Our approach is addressed by roughly three kinds of work: 1) Match BPMN
process to sBPMN, 2) computing likelihoods for next activities based on historic
data and 3) annotating business processes with the generated knowledge.
sBPMN was developed to allow for extending BPMN elements with addi-
tional information and background knowledge to enhance analysis [11, 12]. So
far, existing work already addressed the transformation of BPMN into other
languages like e.g. BPEL [13, 14]. Further work developed ontologies to seman-
tically enrich BPMN in sBPMN [2, 3]. We used an existing ontology developed
by DKM [2].
Another aspect that is tackled in our approach is the computation of the
occurrence of activities in a process in general, but also conditioned on previ-
ous activites, based on historic workflows. Finding predictive models to forecast
activities is in fact a very prominent example in Process Mining. Forecasting
tools exist that on the one hand detect data attributes that influence the choices
in a process [15], as well as to detect decision points and try to minimize un-
certainties in a process [16]. Existing approaches forecast the next activity and
activity durations in a process by using decision trees and rule induction [17, 18],
regression [19] or a classification model to support prediction of activities [20]
and exceptions [21]. Other approaches adressed inferring the future actions of
people from noisy visual input [22]. Approaches use Markov Decision Processes
1
This corresponds to a likelihood of 0.0435
2
This corresponds to a likelihood of 0.0047
Annotating sBPMN Elements with their Likelihood of Occurrence 7
as model and try to maximize the likelihood of the training data under the max-
imum entropy distribution. Surveys exist to give overviews of already addressed
topics in Process Mining [23, 24].
The last addressed topic is the enrichment of the business process with the
generated knowledge. Current work used semantic information to increase the
precision of process models [25]. Tools exist to specify annotations for business
processes [26], as well as web services [27].
5 Conclusions
In this paper we present an approach to map a BPMN process into sBPMN.
We used the workflows of the process to calculate the likelihood of occurence of
each activity and the likelihood of occurence for each next activity, depending on
the current activity. This generated knowledge could in turn be used to enrich
the sBPMN with further information. We used the likelihood of occurence of
the activities to check for stochastically dependecies. In addition, the likelihoods
were used to compute the likelihood of various workflows. Thereby, we assumed
the Markov chain characteristic of the process. Future work includes combining
our previous work of detecting correlations of meta-information and the outcome
of this work. We will combine the detection of correlation of meta-information
and dependency of the workflow to detect crucial parts of the process and detect
unknown influences in the process. We suppose to indicate the critera in decision
points.
References
1. Information technology object management group business pro- cess model and
notation, URL: http://www.iso.org/iso/catalogue detail. htm?csnumber=62652 [ac-
cessed: 2016-10-04].
2. Marco Rospocher, Chiara Ghidini, Luciano Serafini. An ontology for the Business
Process Modelling Notation Formal Ontology in Information Systems - Proceedings
of the Eighth International Conference, FOIS2014, September, 22-25, 2014, Rio de
Janeiro, Brazil, vol. 267, pp. 133-146, IOS Press, 2014
3. Natschlger, Christine: Towards a BPMN 2.0 Ontology, Business Process Model and
Notation: Third International Workshop, BPMN 2011, Lucerne, Swizerland, Novem-
ber 2011
4. J. vom Brocke and M. Rosemann, Eds., BPMN 2.0 for Modeling Business Pro-
cesses, Handbook on Business Process Management 1: Introduction, Methods, and
Information Systems. Springer, 2015, ISBN: 978-3-642-45099-0.
5. W. Yao and A. Kumar, Conflexflow: Integrating flexible clinical pathways into clin-
ical decision support systems using context and rules, Decision Support Systems,
vol. 55, no. 2, 2013, pp. 499515, 1. Analytics and Modeling for Better HealthCare
2. Decision Making in Healthcare.
6. W. R. Gilks, S. Richardson and D. Spiegelhalter. Markov chain Monte Carlo in
practice. CRC press, 1995.
8 Tobias Weller and Maria Maleshkova
7. A. Maaradji, M. Dumas, M. La Rosa and A. Ostovar, Fast and Accu-
rate Business Process Drift Detection, QUT ePrints #83013, QUT, 2015
(http://eprints.qut.edu.au/83013).
8. Tobias Weller, Maria Maleshkova Cognitive Process Designer - An Open-Source
Tool to Capture Processes according to the Linked Data Principles The Semantic
Web: ESWC 2016 Satellite Events, Springer
9. Tobias Weller, Maria Maleshkova Capturing and Annotating Processes using a Col-
laborative Platform Proceedings of the 25th International Conference on World
Wide Web, WWW 2016, ACM, Montral, Canada, April, 2016
10. Max Vlkel, Markus Krtzsch, Denny Vrandecic, Heiko Haller, Rudi Studer Semantic
Wikipedia Proceedings of the 15th international conference on World Wide Web,
WWW 2006, Edinburgh, Scotland, May 23-26, 2006, ACM, Mai, 2006
11. Abramowicz, Witold, Agata Filipowska, Monika Kaczmarek and Tomasz Kacz-
marek. ”Semantically Enhanced Business Process Modeling Notation.” Semantic
Technologies for Business and Information Systems Engineering: Concepts and Ap-
plications. IGI Global, 2012. 259-275. Web. 3 Mar. 2017. doi:10.4018/978-1-60960-
126-3.ch013
12. H. Fernndez Fernndez, Elas Palacios-Gonzlez, Vicente Garca-Daz, B. Cristina
Pelayo G-Bustelo, Oscar Sanjun Martnez, Juan Manuel Cueva Lovelle,
SBPMN An easier business process modeling notation for business users,
Computer Standards & Interfaces, Volume 32, Issues 12, January 2010,
Pages 18-28, ISSN 0920-5489, http://dx.doi.org/10.1016/j.csi.2009.04.006.
(http://www.sciencedirect.com/science/article/pii/S0920548909000300)
13. Guillaume Doux and Frdric Jouault, Jean Bzivin. Transforming BPMN process
models to BPEL process definitions with ATL, In GraBaTs 2009 : 5th International
Workshop on Graph- Based Tools, 2009
14. Saartje Brockmans and Marc Ehrig and Agnes Koschmider and Andreas Oberweis
and Rudi Studer: Semantic Alignment of Business Processes, Proceedings of the
Eighth International Conference on Enterprise Information Systems (ICEIS 2006),
May 2006
15. A. Rozinat, W.M.P. van der Aalst Decision mining in ProM, Proc. of 4th Intl.
Conf. on Business Process Management (BPM’06) (2006), pp. 420425
16. S. Subramaniam, V. Kalogeraki, D. Gunopulos, F. Casati, M. Castellanos, U.
Dayal, M. Sayal. Improving process models by discovering decision points, Infor-
mation Systems, 32 (7) (2007), pp. 10371055
17. C. Apte, S. Weiss. Data mining with decision trees and decision rules. Future
Generation Computer Systems, 13 (1997), pp. 197210
18. Ceci M., Lanotte P.F., Fumarola F., Cavallo D.P., Malerba D. (2014) Completion
Time and Next Activity Prediction of Processes Using Sequential Pattern Mining.
In: Deroski S., Panov P., Kocev D., Todorovski L. (eds) Discovery Science. DS 2014.
Lecture Notes in Computer Science, vol 8777. Springer, Cham
19. Xiao Liu, Zhiwei Ni, Dong Yuan, Yuanchun Jiang, Zhangjun Wu, Jinjun
Chen, Yun Yang, A novel statistical time-series pattern based interval fore-
casting strategy for activity durations in workflow systems, Journal of Systems
and Software, Volume 84, Issue 3, March 2011, Pages 354-376, ISSN 0164-1212,
http://dx.doi.org/10.1016/j.jss.2010.11.927.
20. Ceci, Michelangelo and Lanotte, Pasqua Fabiana and Fumarola, Fabio and Cavallo,
Dario Pietro and Malerba, Donato. Completion Time and Next Activity Prediction
of Processes Using Sequential Pattern Mining, Discovery Science: 17th International
Conference, DS 2014, Bled, Slovenia, October 8-10, 2014. Proceedings, 2014, 49–61,
http://dx.doi.org/10.1007/978-3-319-11812-3 5
Annotating sBPMN Elements with their Likelihood of Occurrence 9
21. Y. Hai-tao, D. Bin and S. Zheng-xiao, ”Workflow Exception Forecast-
ing Method Based on SVM Theory,” 2008 International Symposium
on Computational Intelligence and Design, Wuhan, 2008, pp. 81-86.,
http://dx.doi.org/10.1109/ISCID.2008.66
22. Kitani K.M., Ziebart B.D., Bagnell J.A., Hebert M. (2012) Activity Forecasting. In:
Fitzgibbon A., Lazebnik S., Perona P., Sato Y., Schmid C. (eds) Computer Vision
ECCV 2012. ECCV 2012. Lecture Notes in Computer Science, vol 7575. Springer,
Berlin, Heidelberg
23. Indranil Bose, Radha K. Mahapatra, Business data mining a machine learning
perspective, Information & Management, Volume 39, Issue 3, 20 December 2001,
Pages 211-225, ISSN 0378-7206, http://dx.doi.org/10.1016/S0378-7206(01)00091-X.
24. zur Mühlen, Michael and Shapiro, Robert. Business Process Analytics, Handbook
on Business Process Management 2: Strategic Alignment, Governance, People and
Culture, 2010, Berlin, 137–157, http://dx.doi.org/10.1007/978-3-642-01982-1 7
25. Matthias Born, Florian Drr, and Ingo Weber. 2007. User-friendly semantic an-
notation in business process modeling. In Proceedings of the 2007 international
conference on Web information systems engineering (WISE’07), Mathias Weske,
Mohand-Sad Hacid, and Claude Godart (Eds.). Springer-Verlag, Berlin, Heidelberg,
260-271.
26. K. Hinge, A. Ghose and G. Koliadis, ”Process SEER: A Tool for Semantic Ef-
fect Annotation of Business Process Models,” 2009 IEEE International Enter-
prise Distributed Object Computing Conference, Auckland, 2009, pp. 54-63. doi:
10.1109/EDOC.2009.24
27. Andreas He, Eddie Johnston, and Nicholas Kushmerick. 2004. ASSAM: a tool for
semi-automatically annotating semantic web services. In Proceedings of the 3rd
International Conference on Semantic Web Conference (LNCS-ISWC’04), Sheila A.
McIlraith, Dimitris Plexousaki, and Frank Van Harmelen (Eds.). Springer-Verlag,
Berlin, Heidelberg, 320-334. DOI: http://dx.doi.org/10.1007/978-3-540-30475-3 23