Configurable Process Mining: Semantic Variability in Event Logs

Aicha Khannat

, Hanae Sbai

and Laila Kjiri

AlQualsadi, Research Team ENSIAS, Mohammed V University of Rabat, Rabat, Morocco

FST Mohammedia, University Hassan II of Casablanca, Casablanca, Morocco

Keywords: Configurable Process Model Discovery, Process Mining, Event Log, Ontology, Semantic.

Abstract: Configurable process model represents a reference model regrouping multiple business process variants. The

configurable process models offer various benefits like reusability and more flexibility when compared to

business process models. The challenges encountered while managing this type of models are related to the

creation and the configuration. Recently, process mining offers techniques to discover, check conformance of

models, and enhance configurable process models using a collection of event logs, that captures traces during

the execution of process variants. However, existing works in configurable process discovery lack the

incorporation of semantics in the resulting model. Historically, semantic process mining has been applied to

event logs to improve process discovery with respect to semantic. Furthermore, from the best of our

knowledge, configurable process mining approaches do not fully support semantics. In this paper, we propose

a novel method to enrich the collection of event logs with configurable process ontology concepts by

introducing semantic annotations that capture variability of elements present in the logs. This is a first step

towards discovering a semantically enriched configurable process.

1 INTRODUCTION

Configurable Process models regroup multiple

process behaviours into a single model with the

possibility to be configured according to the needs of

an execution environment. Each one of the resulting

business process models, called process variant,

captures a specific behaviour of the reference model

(Derguech, 2017).

Configurable process models seem to be useful

for large organizations that manage similar processes

in different conditions like insurance companies,

banks, and universities (Benítez, 2017).

Configurable process models offer also multiple

advantages such as guaranteeing consistency between

business process models, avoiding business process

clones (De Medeiros, 2008) and offering a certain

degree of flexibility regarding the possible ways to

execute the process (Benítez, 2017). They are

constructed using two methods: i) manual approach:

which preconizes merging multiple process variants

from scratch (La Rosa, 2013); (Derguech, 2011);

(Assy, 2013), ii) automatic approach: which is based

on the application of mining techniques (Buijs, 2013).

Concerning the manual approach, since the

variability is identified in a specific domain, designer

collects different process variants which will be

merged into one model and represented by one of the

existing configurable process modeling languages

like C-EPC (La Rosa, 2011); C-BPMN (Rosa, 2017);

C-YAWL (Gottschalk, 2008) and EVR-BPMN (Sbai,

2015). Contrary to automatic approach (Buijs, 2013),

where configurable process models are created

directly from real time recorded data of a collection

of event logs. There are three main process mining

axes:

 Discovery: creation of configurable process

models using a collection of event logs,

 Conformance: analysis of configurable

process models regarding a collection of event

logs,

 Enhancement: improvement of configurable

process models using data captured in

collection of event logs.

The use of event logs for business process mining,

redresses the problem of having limited information

about the way of working in organizations (Detro,

2017). However, business process mining based on

the real-life logs, have some weaknesses: i)

production of large and spaghetti-like models, ii)

production of models with low fitness and iii)

production of models with low precision or low

768

Khannat, A., Sbai, H. and Kjiri, L.

Conﬁgurable Process Mining: Semantic Variability in Event Logs.

DOI: 10.5220/0010484207680775

In Proceedings of the 23rd International Conference on Enterprise Information Systems (ICEIS 2021) - Volume 1, pages 768-775

ISBN: 978-989-758-509-8

generalization (Augusto, 2018). Hence, to improve

the quality degree of process models discovered, it is

crucial to start with high-level event logs. For that

reason, existing approaches introduced semantic

processing, based on ontologies, to enhance the

quality of event logs and ensure that the events

present in the event log directly correspond to the

activities that are recognizable for process

stakeholders. Then, the analysis made based on event

logs data will be more accurate and correct compared

to syntactic analysis.

Similarly, to business process mining, existing

approaches in the configurable process mining field

use domain ontology to unify event logs data. As the

analysis of configurable process models are based on

a collection of event logs, the challenge is to consider

the variability expressed in the collection of event

logs and make sure to capture the same in the

resulting configurable process model.

Despite of efforts that has been made to introduce

semantic in process discovery, configurable process

discovery approaches are limited to the syntactic

level. The challenge with the syntactic analysis is the

dependence on labels presented in the event logs and

this causes a lack of the abstraction level required for

real world applications (Okoye, 2020). To overcome

this, few papers combine semantic concepts with

process mining techniques to provide semantic

analysis in a high level of abstraction (Detro, 2017).

The incorporation of semantics in the configurable

process models can help to exchange process

information between the applications in the most

efficient manner (Detro, 2017).

Many works in the context of manual approach,

propose the integration of semantics for managing

configurable process models and their customization

(Detro, 2017); (Benítez, 2017); (Buijs, 2013).

However, for the automatic approach, in which the

configurable process is discovered from event logs,

they focus on the discovery of variable fragments and

shared fragments to derive the configurable process

model without including semantic concepts in the

final model. As well, the semantic technologies are

used to reduce the complexity of the configurable

process model or to give assistance during process

configuration without the inclusion of semantics in

the configurable process model.

According to our previous work (Khannat, 2020),

we proposed a framework to discover semantically

enriched configurable process models based on a

collection of semantically enriched event logs.

In this paper, we propose an approach to enrich

the collection of event logs with variability concepts

and domain ontology as part of the event logs pre-

processing component. The objective is to prepare the

collection of event logs using ontologies as a first step

towards discovering a semantically enriched

configurable process model.

The remainder of this paper is structured as

follows: Section 2 describes the main concepts related

to our work. Section 3 provides main ideas of related

works regarding semantic enrichment of event logs.

Section 4 presents an overview of the proposed

approach. Finally, Section 5 concludes this paper and

discusses future work.

2 BASIC CONCEPTS

In this section, we present two main concepts related

to our work that are configurable process model

discovery and semantic in the event log.

2.1 Configurable Process Model

Discovery

Configurable process model is a process model that

describes both the commonalities shared by all

process variants and their differences (Derguech,

2017). Common parts are presented in all process

variants, while variable parts represent options that

can be configured depending on the process execution

context. Process Mining techniques are used to

automatically discover configurable process models

based on collection of event logs.

Figure 1 (Buijs, 2013) illustrates existing

approaches for automatic discovery of configurable

process models.

Figure 1: Configurable process model discovery

approaches.

 Approach 1: Mining of the process variant

corresponding to each event log and merging

models, then discovering its configurations.

Conﬁgurable Process Mining: Semantic Variability in Event Logs

769

 Approach 2: Merging event logs and

discovering common parts then extracting

process variants and merging them to obtain the

configurable process model.

 Approach 3: Merging event logs and mining

configurable process model, then discovering

the configurations.

 Approach 4: Discovering the configurable

process model and its configurations at the

same time.

The approach 4 is proposed to overcome

challenges of other approaches, the configurable

process model is smaller and simpler compared to

other models.

The quality of the configurable process model

directly impacts the customization and the extraction

of process variants (Detro, 2017), the more

comprehensive is the model, the easier will be the

customization. Thus, enriching configurable process

model with semantics improves the representation of

processes and allows automation of configuration

task with more flexibility and adaptation to different

business contexts (El Faquih, 2020). Semantics

consist of the integration of ontologies during the

process creation phase or the process analysis phase.

Ontology is defined as a set of concepts and existing

relationships between them in formalized

representation (Detro, 2020). Introduction of

ontologies enables sharing knowledge, unifying

vocabularies, and adjusting the level of details. Two

main ontologies are used in the field of configurable

process models: i) domain ontology: regroups

concepts that belong to specific domain, and ii)

variability ontology: captures the variability of the

process variation points. Some existing approaches

use these ontologies for two main purposes: i)

configuration: derives rules that assist users during

the configuration process, and ii) validation: ensures

semantic correctness of the process variants.

Existing approaches use Semantic Business

Process Mining techniques to perform analysis on

process execution traces at the conceptual level, this

enables deriving knowledge from event logs. Thus,

the stage related to preparation of event logs is crucial

in process mining, specifically in semantic

configurable process mining.

2.2 Semantics in Event Logs

Event logs resume information about the process

execution, such timestamp, case, activity, and

resources (Allani, 2016). These real data are

considered with great importance in the field of

process mining, as they allow discovering,

conforming, and enhancing business process models.

There are two formats to represent and store event

logs: MXML (Mining eXtensible Markup Language)

and XES (eXtensible Event Stream) (Verbeek, 2010).

Both formats define an event log as a sequence of

events but using different concepts and attributes.

MXML uses the below concepts to describe

process execution traces (Günther, 2006):

 WorkflowLog: represents a log file.

 Process: regroups events having been occurred

during the execution of a specific process.

 ProcessInstance: represents single execution of

the process.

 Data: represents Data attributes that can be

associated to each element of the log.

 AuditTrailEntry: describes one event in the log

and contains the below child elements:

 WorkflowModelElement: captures the

activity name that triggered the event.

 EventType: captures the type of the

event (e.g. start, complete).

 Originator: captures the resource name

that executed the activity.

 Timestamp: captures the time at which

the event occurred in the system.

XES uses also specific concepts to represent event

log data (Verbeek, 2010):

 Log: corresponds to Workflow in MXML.

 Extension: specifies semantics of an attribute,

which could be either a standard extension or

some user-deﬁned extension.

 Trace: matches to ProcessInstance in MXML.

 Classifier: assigns an identity to each event.

 Attribute: stores data about each element of

Regarding expression of semantics in event logs,

the two formats store semantic annotations in

different ways:

 Case of MXML event log: New format SA-

MXML (Semantically Annotated MXML) has

been defined to represent MXML event logs

enriched semantically. The SA-MXML format

is an extension of the MXML format whereby

all elements (except for AuditTrailEntry and

Timestamp) have an optional extra attribute

called modelReference that links to a list of

concepts in ontologies (the concepts are

expressed as URIs) (De Medeiros, 2008).

 Case of XES event log: XES uses the

extension Semantic to support semantic

annotations that refers to ontology concepts.

This is inserted as an attribute in all levels (log,

trace, event and meta) of type ‘string’ with key

ICEIS 2021 - 23rd International Conference on Enterprise Information Systems

770

‘modelReference’ and value that reference to

model concepts in an ontology.

Both formats SA-MXML and XES are supported by

ProM framework in process mining applications.

3 RELATED WORK

In this section, we present existing works on semantic

enrichment of event logs.

The approach presented in (Okoye, 2020)

introduces semantic annotations to link the event log

to the domain ontology in order to answer some

questions with regards to different learning

patterns/behaviour and discover unobserved learning

behaviours or patterns. They argue that the analysis

provided by process mining techniques can be

improved using semantics. The authors in (Cairns,

2014) propose a (semi)automatic procedure to link

training labels of the educational event log to the right

concepts of a training ontology, in order to generate

and analyze a less complex process model. The work

(Yongsiriwit, 2017) proposes to semantically

represent event logs using the extended ontology

NCFO (Neighborhood Context Fragment Ontology)

in order to compare event logs to an under-design

process for assisting business process variants design.

Authors in (Nykänen, 2015) defined two main

ontologies to be associated with an event log: i)

process ontology: describes activities of the target

process model and relationships between them and ii)

product ontology: describes the object (resource)

used by the process. The main purpose of enriching

process mining using events logs with associated

ontology structures is to analyze the process models in

different abstraction levels, which greatly helps to

understand complicated processes. (Jareevongpiboon,

2013) introduces a methodology to combine domain

ontology, company-specific ontologies, and

databases to obtain multiple levels of abstraction for

mining and analysis. They propose to map concepts

from ontologies to process execution data for

improvement of results in process mining and

analysis. The process discovered can be viewed in

two ways: i) it can be viewed at the domain concepts

level and ii) it can be viewed at a company speciﬁc

level. The application of this methodology proves that

semantics enhance the business object dimension of

analysis. The authors of (Detro, 2017) propose an

approach to explore event logs data using domain

ontology and variation points ontology with the

objective of giving suggestions during configurable

process model customization. The work (Sellami,

2012) takes interest of the organizational perspective,

it presents an approach to semantically annotate event

log with organizational ontology, which allows

creating a knowledge base related to the relationship

between performers in a workflow.

To sum up, Table 1 shows a comparison of the

related works, presented in this section, according to

the following criteria:

 Event Log Category: indicates if the approach

uses a single event log or a collection of logs.

 Event Log Language: specifies the language

used to represent the event logs.

 Element Annotated: indicates the element

annotated using the ontology (task or resource).

 Type of Ontologies: determines the type of

ontology used.

 Ontology Language: determines the language

used to represent the ontologies.

 Objectives: identifies the objectives for

semantic enrichment of traces.

The comparison of these works shows us that few

approaches are interested in semantic enrichment of

event logs collection. The works (Detro, 2017) and

(Yongsiriwit, 2017) enrich the collection of event

logs to extract process variants. The other approaches

(Okoye, 2020); (Cairns, 2014); (Nykänen, 2015);

(Jareevongpiboon, 2013); (Sellami, 2012) are limited

to the enrichment of a single event log. Thus, the

existing approaches are not sufficient for the

preparation of the collection of event logs with the

purpose of mining configurable process models. In

addition, most of the works are limited to activity

elements in the semantic annotation. Few works

(Nykänen, 2015); (Jareevongpiboon, 2013); (Sellami,

2012) that propose semantic annotation for resource

element. So, there is a need to integrate all

perspectives when semantically enriching event logs.

As well, the existing approaches (Okoye, 2020);

(Cairns, 2014); (Sellami, 2012) use domain

ontologies, and this presents a lack for the discovery

of configurable process models, knowing that this

type of models should manage variability.

Moreover, approaches that handle with collection

of event logs (Yongsiriwit, 2017); (Detro, 2017) are

using semantics for configuration only and the

configurable process model extracted is not enriched

with semantics.

The existing works use OWL (Web Ontology

Language) and WSML (Web Service Modeling

Language) to represent ontologies and use event logs

expressed in XES (eXtensible Event Stream) or

MXML (XML-based user interface markup

language).

Conﬁgurable Process Mining: Semantic Variability in Event Logs

771

Table 1: Summary of approaches related to semantic enrichment of event logs.

Work

Event log

category

Event log

language

Element

annotated

Type of ontologies

Ontology

language

Objectives

(Okoye,

2020)

Single -- Task Domain ontology OWL

Process discovery

and enhancement

(Cairns,

2014)

Single MXML Task Domain ontology WSML Process discovery

(Yongsiri

wit, 2017)

Collection XES Task

Process model

ontology

OWL

Assisting business

process variants

design

(Nykänen,

2015)

Single --

Task

Resource

Process model

ontology

Domain ontology

OWL

Analysis of process

models in different

abstraction levels

(Jareevon

gpiboon,

2013)

Single MXML

Task

Resource

Domain ontology

WSML

Process discovery

and enhancement

(Detro,

2017)

Collection MXML Task

Domain ontology

Variability ontology

OWL

Automatic

suggestions during

process

configuration

(Sellami,

2012)

Single XES Resource Domain ontology OWL

Discovery of

relationship between

performers in a

workflow

When analyzing the existing approaches, we

deduce that most of them are limited to semantic

annotation of activity element with domain ontology

and apply their approaches to single event log. These

approaches seem to be not suitable for the preparation

of event logs collection in the field of configurable

process models, as we need to enrich the collection of

event logs with domain concepts and variability

concepts. For these reasons, we propose an approach

to semantically annotate collection of event logs with

configurable process model ontology and domain

ontology. Then, process mining techniques will be

applied on the enriched event logs to discover

semantically annotated configurable process models.

4 APPROACH OVERVIEW

In our previous work (Khannat, 2020), we proposed

the framework for discovering semantically enriched

configurable processes. Fig 2 represents a simplified

illustration of the proposed framework.

Event Logs Pre-processing Component. This

component is used to merge the collection of event

logs and prepare the resulting log by adding semantic

annotations that link event logs elements to concepts

formalized in two ontologies: domain ontology and

configurable process model ontology (CPMO).

Configurable Process Model Discovery. This

component takes as input the event log prepared in

the first component and applies process mining

techniques to discover the configurable process

model, that is enriched semantically with the same

ontologies as the event log prepared, and the

appropriate rules of configuration.

Figure 2: Framework for automatic discovery of

semantically enriched configurable process model.

The main idea is to semantically enrich collection

of event logs as this will lead to enhance the quality

of the discovered model and gives analyst views in

multiple abstraction levels, it will also improve the

quality of process variants extracted from the

configurable process model, knowing that they will

be enriched and validated semantically. We propose

to enrich event logs using two ontologies, the first one

to express variability in event logs and the second one

to link event log to domain concepts. This step will

allow creation of high-level annotated event logs that

ICEIS 2021 - 23rd International Conference on Enterprise Information Systems

772

will optimize process mining application and allow

semantic validation of process elements before being

discovered. In this paper, we focus on the annotation

of event logs using the CPMO. We suppose that the

collection of event logs is already constructed and

contains only instances of similar process variants.

The corresponding variables fragments are supposed

already identified based on existing methods (Sikal,

2018); (Vaca, 2019). Regarding variability

perspectives, we are interested in activities and

resources and we consider that the variability of

resources depends on the variability of activities.

Figure 3 illustrates the proposed approach for

enriching collection of event logs using CPMO.

Figure 3: Approach overview of semantic enrichment of

event logs.

To achieve our objective of integrating semantic

annotations that link activities and resources in event

logs to variability concepts formalized in CPMO, we

propose to follow the below steps:

Step 1: Merging the event logs into one

consolidate event log: The merging technique is

similar to the techniques presented in the approach

(Suriadi, 2017). We will create one consolidated

event log in the same format as the input event logs,

this file will contain all instances included in starting

logs. Each element of types: process, process instance

and activity will have a unique identifier (e.g. (i), (i,

j), (i, j, k)) that links the element to its parent element

in the new event log.

Step 2: Defining semantic annotations to use: In

this step, we define the semantic annotations to be

integrated in the event log, resulting from the merge

of the collection. These semantic annotations will

refer to concepts of CPMO.

Table 2 (El Faquih, 2020) depicts the CPMO

variability classes, subclasses, and relationships.

Table 2: CPMO concepts.

The main idea is to insert new attributes, in the event

log merged, that will contain values of the CPMO

classes. Table 3 presents the attributes that will be

inserted in the event log, their values, and the

concerned elements in the case of MXML event log

or XES event log.

We use the notation CPMO#ontology_concept

while referring to a concept in the CPMO.

Step 3: Marking variability into the event log:

This step is about including the defined attributes into

the log based on the variability specification file. To

achieve this, we propose the algorithm, illustrated in

Figure 4, which takes as input the event log in MXML

Table 3: Attributes used for semantic annotation with CPMO.

Attributes Signification Possibles values

Targeted element in

MXML event log

Targeted

element in XES

event log

VariabilityType

The type of

variability

CPMO#variable

Process / ProcessInstance

Trace

CPMO#variationPoint AuditTrailEntry Event

CPMO#variant AuditTrailEntry Event

VarPtType

The type of

variation point

CPMO#alternative

CPMO#optional

CPMO#optionalAlternative

AuditTrailEntry Event

VarType

The type of variant

CPMO#default

CPMO#variant

AuditTrailEntry Event

CPMO

variability

classes

Subclasses Relationships

Variable ------ CPM contains variable

Variation_

point

Alternative

Optional

Optional_

alternative

CPM contains variation

point

Variation_point is_a

alternative

Variation_point is_a

optional

Variation_point is_a

optional _alternative

Variation_point

has_variant variant

Variant Default

CPM contains variant

Variant has default

Conﬁgurable Process Mining: Semantic Variability in Event Logs

773

format and the variability specification file and

generates as output the semantically annotated event

log using CPM ontology. The same algorithm can be

adapted to XES event logs format.

Figure 4: Algorithm proposed to enrich MXML event log

with CPMO concepts.

Figure 5: Fragment of SA-MXML file enriched with

CPMO.

Through application of the algorithm proposed, we

can generate semantically enriched event log that

links variable elements to CPMO concepts. Figure 5

represents an extract from the resulting merged and

annotated event log in SA-MXML format.

5 CONCLUSIONS AND FUTURE

WORK

Configurable process mining still confronting

challenges related to variability and complexity of the

discovered models. Semantics represent a great key

to enhance configurable process model quality,

however, the application of semantic techniques still

limited to validation or configuration. Thus, our

framework aims to integrate semantics in the

discovery of configurable process models to manage

variability more easily and give a conceptual view of

the model. The proposed framework is based on two

ontologies: Domain ontology and CPMO. In this

paper, we proposed an algorithm to semantically

enrich collection of event logs using CPMO. The

resulting event log will be enriched with domain

ontology and then used as input for configurable

process mining techniques to discover semantically

enriched configurable process model.

As future work, we aim to complete the first

component implementation by annotating the event

log with domain ontology and validate this

component by presenting a use case application.

Moreover, we will focus on the application of process

mining techniques on the event log prepared to

discover the configurable process model.

REFERENCES

Allani, O., & Ghannouchi, S. A. (2016). Verification of

BPMN 2.0 process models: an event log-based

approach. Procedia Computer Science, 100, 1064-

1070.

Arriagada-Benítez, M., Sepúlveda, M., Munoz-Gama, J., &

Buijs, J. C. (2017). Strategies to automatically derive a

process model from a configurable process model based

on event data. Applied sciences, 7(10), 1023.

Assy, N., Chan, N. N., & Gaaloul, W. (2013, June).

Assisting business process design with configurable

process fragments. In 2013 IEEE International

Conference on Services Computing (pp. 535-542).

IEEE.

Augusto, A., Conforti, R., Dumas, M., La Rosa, M., Maggi,

F. M., Marrella, A., ... & Soo, A. (2018). Automated

discovery of process models from event logs: Review

ICEIS 2021 - 23rd International Conference on Enterprise Information Systems

774

and benchmark. IEEE transactions on knowledge and

data engineering, 31(4), 686-705.

Buijs, J. C., van Dongen, B. F., & van der Aalst, W. M.

(2013). Mining configurable process models from

collections of event logs. In Business process

management (pp. 33-48). Springer, Berlin, Heidelberg.

Cairns, A. H., Ondo, J. A., Gueni, B., Fhima, M.,

Schwarcfeld, M., Joubert, C., & Khelifa, N. (2014,

November). Using semantic lifting for improving

educational process models discovery and analysis.

In SIMPDA (pp. 150-161).

De Medeiros, A. K. A., Van der Aalst, W., & Pedrinaci, C.

(2008). Semantic process mining tools: Core building

blocks.

Derguech, W., & Bhiri, S. (2011, June). Merging business

process variants. In International Conference on

Business Information Systems (pp. 86-97). Springer,

Berlin, Heidelberg.

Derguech, W., Bhiri, S., & Curry, E. (2017). Designing

business capability-aware configurable process

models. Information Systems, 72, 77-94.

Detro, S. P., Santos, E. A. P., Panetto, H., Loures, E. D. F.

R., & Lezoche, M. (2017, September). Managing

business process variability through process mining and

semantic reasoning: An application in healthcare.

In Working Conference on Virtual Enterprises (pp.

333-340). Springer, Cham.

Detro, S. P., Santos, E. A. P., Panetto, H., Loures, E. D.,

Lezoche, M., & Cabral Moro Barra, C. (2020).

Applying process mining and semantic reasoning for

process model customisation in healthcare. Enterprise

Information Systems, 14(7), 983-1009.

El Faquih, L., & Fredj, M. (2020). Ontology-Based

Framework for Quality in Configurable Process

Models. In Sustainable Business: Concepts,

Methodologies, Tools, and Applications (pp. 464-478).

IGI Global.

Gottschalk, F., Van Der Aalst, W. M., Jansen-Vullers, M.

H., & La Rosa, M. (2008). Configurable workflow

models. International Journal of Cooperative

Information Systems, 17(02), 177-221.

Günther, C. W., & van der Aalst, W. M. (2006, September).

A generic import framework for process event logs.

In International Conference on Business Process

Management (pp. 81-92). Springer, Berlin, Heidelberg.

Jareevongpiboon, W., & Janecek, P. (2013). Ontological

approach to enhance results of business process mining

and analysis. Business Process Management Journal.

Khannat, A., Sbai, H., & Kjiri, L. (2020, December). Event

Logs Pre-processing for Configurable Process

discovery: Ontology-Based Approach. In 2020 6Th

International IEEE congress on information science

and technology (CiSt). IEEE, in press.

La Rosa, M., Dumas, M., Ter Hofstede, A. H., & Mendling,

J. (2011). Configurable multi-perspective business

process models. Information Systems, 36(2), 313-340.

La Rosa, M., Dumas, M., Uba, R., & Dijkman, R. (2013).

Business process model merging: An approach to

business process consolidation. ACM Transactions on

Software Engineering and Methodology

(TOSEM), 22(2), 1-42.

Nykänen, O., Rivero-Rodriguez, A., Pileggi, P., Ranta, P.

A., Kailanto, M., & Koro, J. (2015, September).

Associating event logs with ontologies for semantic

process mining and analysis. In Proceedings of the 19th

International Academic Mindtrek Conference (pp. 138-

143).

Okoye, K., Islam, S., Naeem, U., & Sharif, S. (2020).

Semantic-based process mining technique for

annotation and modelling of domain processes.

International Journal of Innovative Computing,

Information and Control, 16(3), 899-921.

Rosa, M. L., Aalst, W. M. V. D., Dumas, M., & Milani, F.

P. (2017). Business process variability modeling: A

survey. ACM Computing Surveys (CSUR), 50(1), 1-45.

Sbai, H. (2015). PAIS (process aware information systems)

orienté services : modélisation et évolution des

processus configurables (Doctoral dissertation,

ENSIAS, University of Rabat, Morocco).

Sellami, R., Gaaloul, W., & Moalla, S. (2012, June). An

ontology for workflow organizational model mining.

In 2012 IEEE 21st International Workshop on Enabling

Technologies: Infrastructure for Collaborative

Enterprises (pp. 199-204). IEEE.

Sikal, R., Sbai, H., & Kjiri, L. (2018, October).

Configurable process mining: variability Discovery

Approach. In 2018 IEEE 5th International Congress on

Information Science and Technology (CiSt) (pp. 137-

142). IEEE.

Suriadi, S., Andrews, R., ter Hofstede, A. H., & Wynn, M.

T. (2017). Event log imperfection patterns for process

mining: Towards a systematic approach to cleaning

event logs. Information Systems, 64, 132-150.

Varela-Vaca, Á. J., Galindo, J. A., Ramos-Gutiérrez, B.,

Gómez-López, M. T., & Benavides, D. (2019,

September). Process mining to unleash variability

management: discovering configuration workflows

using logs. In Proceedings of the 23rd International

Systems and Software Product Line Conference-

Volume A (pp. 265-276).

Verbeek, H. M. W., Buijs, J. C., Van Dongen, B. F., & Van

Der Aalst, W. M. (2010, June). Xes, xesame, and prom

6. In International Conference on Advanced

Information Systems Engineering (pp. 60-75).

Springer, Berlin, Heidelberg.

Yongsiriwit, K. (2017). Modeling and mining business

process variants in cloud environments (Doctoral

dissertation, Université Paris-Saclay (ComUE)).

Conﬁgurable Process Mining: Semantic Variability in Event Logs

775