E-learning represents a significant opportunity to advance educational equity, integrating contemporary pedagogical approaches with sophisticated information technology. As the demand for e-learning intensifies, optimizing and enhancing e-learning processes assumes greater importance to guarantee a high-quality learning experience. Our approach entails the utilization of process mining techniques to conduct a comprehensive and in-depth analysis of e-learning workflows, coupled with the application of machine learning for the purpose of implementing enhancements and optimizing outcomes. In the first layer of our approach, process mining techniques such as Disco and Inductive Miner are used in educational process mining because they are specifically designed for visualizing and analyzing process data. These tools are integral to educational process mining as they focus on discovering, monitoring, and improving e-learning processes, which makes them more suited for this domain than tools like Celonis. Celonis, although powerful, is not tailored for educational contexts and doesn’t offer the specific features needed for analyzing educational workflows. In the second layer of our approach, machine learning techniques such as Decision Tree, Linear Regression, Logistic Regression, Random Forest, and SVM algorithms are used due to their ability to effectively handle classification, prediction, and regression tasks. These algorithms enable accurate predictions of learner behavior and performance, which helps optimize the learning experience and personalize learning paths.

According to the Global Health Observatory, stroke ranks second worldwide in causing dementia, right after Alzheimer’s disease. The mortality rate linked to dementia resulting from stroke is high because symptoms are often recognized late, and stroke can be misinterpreted as other brain disorders. Early detection and diagnosis of stroke is crucial. Therefore, increasing awareness of stroke symptoms and implementing preventive measures becomes imperative. Prompt intervention by healthcare professionals can improve outcomes and reduce long-term complications of stroke. The research introduces an innovative approach for early stroke prediction using a fuzzy scoring-based Decision Support System. This approach encompasses three main modules: Mind map-based Data Modeling, Fuzzy scoring computing, and Machine Learning (ML)-Based Decision System. By incorporating fuzzy logic, the approach extracts valuable knowledge from imprecise and uncertain data. Combining the fuzzy stroke risk model with a ML-based decision support system aims to enhance stroke prediction accuracy and improve preventive measures and patient outcomes. The approach’s effectiveness was validated using real clinical data and tested with various ML classifiers, including K-Nearest Neighbor (KNN), Logistic Regression (LR), Decision Tree (DT), Artificial Neural Network (ANN), and Support Vector Machine (SVM). The results showed a strong correlation between stroke cases and computed risk-scoring values. In comparison to predictions without fuzzy scoring and other related works, the stroke risk prediction using the proposed approach demonstrated higher accuracy, making it a promising method for early stroke detection and prevention.

With today’s technological advancements, the medical field generates a massive number of electronic medical records (EMRs) daily. These EMRs contain critical data that is primarily accessible and interpretable by healthcare experts. However, their unstructured nature presents significant challenges for analysis and interpretation. Transforming these unstructured documents into a structured format is crucial for scientific research and poses a complex challenge for data scientists. This paper presents an approach for converting unstructured textual documents into structured databases, demonstrated through a case study involving the processing of Word medical expert reports into structured data within Excel spreadsheets. These expert assessments, conducted on behalf of the National Health Insurance Fund (known as CNAM), covered various patient requests, ranging from prescribed rest to prolonged absences from work due to long-term illness or permanent incapacity. This method has the potential to significantly enhance healthcare decision-making efficiency and pave the way for further developments in medical informatics.

The emergence of Intelligent BPM (IBPM) is built upon adding smart technologies and business intelligence to BPM. In such a harsh environment, improving BPs using the most appropriate and adequate technologies is considered to be a complex task to deal with. Recently, for the sake of exploring the use of technologies 4.0 in different domains using specific languages and in order to efficiently improve BPs in the context of IBPM, we need to define a specific approach for deciding about the technologies 4.0 to adopt. Thereby, such an approach would improve business processes effectively in this context of IBPM. For this purpose, it is worth noting that considering performance criteria can help us to integrate a specific intelligent technology to a given business process and it is important to develop an IBPM ontology as a first contribution in our approach. This approach was implemented in a tool named BPIGuide. To evaluate this approach, we have studied the post-operative monitoring (POM) process. The results of this study would help users to select the most adequate technology 4.0 that best fit their needs on performance criteria.

In Technology-Enhanced-Learning (TEL) field, Learning Design (LD) research aims to support educators in assisting their learners in achieving the Intended Learning Objectives (ILOs) faster and more effectively. We focus on non-technical savvy educators that are aiming for efficient flipped learning scenarios, including Agile methodology, and providing a gamified learning environment to influence learners' engagement and motivation. However, to the best of our knowledge, no pedagogical standards support the description of this type of scenario, and no authoring tools assist in its creation. Therefore, in this paper, we introduce the implementation of (i) an authoring tool that is based on (ii) IMS LD extension approach. After the examination of a group of well-known authoring tools, we specify the requirements to which it should respond. Moreover, we present its evaluation according to LD experts’ criteria.

Agility is a contemporary approach to IT project management, which also can be used in education. Students learn through the gradual implementation of iterative projects with information exchange between team members. Agility is above all a mindset, being agile is quite simply being able to adapt to an environment that you know perfectly well, but which changes and is metamorphosed more or less slowly and which is never immutable and frozen. Various researches focused on the assessment of innovative teaching methods to promote the acquisition of professional skills (e.g. project-based learning, active and collaborative learning, etc.). In addition, adaptive learning is a pedagogical method favoring tailor-made eLearning to respond to the acquisition of certain skills, through the adaptation of pedagogical resources according to the learners’ needs. In this paper, we propose a new learning process essentially based on the four fundamental values of the agile manifesto. It is a challenge raised to attain the maximum organizational and strategic advantages with the implementation of these four agile values in secondary education.

In agile, students learn through the gradual implementation of iterative projects with information exchanges between team members. Various research works focused on the assessment of innovative teaching methods to promote the acquisition of professional skills (e.g. project-based learning, active and collaborative learning, flipped classes, digital assisted training, etc.). In addition, adaptive learning is a pedagogical method favoring tailor-made e-learning to respond to the acquisition of certain skills, through the adaptation of pedagogical resources according to the learners’ needs. Therefore, to establish a model based on these different methods in order to benefit from their advantages, we based our work on the BPM (Business Process Management) approach, which constitutes the means of implementing the desired agility in the learning process. Thanks to such a cyclical and continuous improvement approach, the learning process will evolve and take into account not only new needs but also the specificities of each involved actor (learner or teacher).

Traditional Business Process Management (BPM) should take advantage from the innovations brought by technologies such as Social Media, Internet of Things, Big Data Analytics, Artificial Intelligence, Blockchain, Augmented Reality, Additive Manufacturing, etc. This leads to the emergence of Intelligent BPM (iBPM) which consists of adding smart technologies and business intelligence to BPM. Selecting the most appropriate and adequate technologies is a complex task in such a harsh environment. Thereby, we propose an approach for selecting the most suitable technologies 4.0 in order to improve business processes effectively in this context of iBPM. For this purpose, it is worth noting that considering performance criteria can help us to integrate a specific intelligent technology to a given business process and it is important to develop an iBPM ontology as a basic step of our approach.

Metrics are advantageous for all software development companies to assess and ensure quality. Therefore, integrating metrics is essential throughout the software development process to track, and correct deviations, and improve continuously. In this paper, we examine how to integrate quality assessment through metrics into the Scrum development process through a new approach entitled Metrics for Quality in Scrum (MQScrum). The MQScrum approach makes a valuable contribution to track Scrum projects even if using agile project management tools without slowing teams down. So, it facilitates the teams monitoring quality, discovering problems, and making the appropriate improvements thanks to the MQScrum reports. Firstly, we introduce a Scrum meta-model enhanced by metrics-related concepts. Then, we represent a layered model for our MQScrum approach illustrating the data processing across three layers from the Scrum process layer to the layers of Scrum concepts and finally to that of metrics. Consequently, the proposed solution is an alignment of the Scrum process with a metric-based Scrum system. Finally, it demonstrates process tasks from which pertinent data is collected and then stored in a database based on Scrum and metrics related concepts.

Collaborative e-learning based on several means of electronic communication mechanism specifically internet and web2.0 (such as forum, virtual classroom, videoconference, etc.) has become the current trend in e-learning systems.The cloud computing environment presents itself an important infrastructure for supporting these collaborative e-learning systems and for optimizing these service-based systems in a multi-tenant way.In this paper, we discuss the problem of discovering multi-tenant e-learning processes.For this, we propose an architecture for a configurable learning process discovery environment and the allocation of resources (Saas, Paas, Iaas) necessary for the execution of the corresponding learning activities.

Cloud computing has a model for utility computing that promotes on-demand scalability, flexible application deployment and reuse. It also promotes multi-tenancy for efficient resource utilization by sharing hardware and software infrastructure among multiple users. Multi-tenant applications running on a cloud infrastructure are provided to clients as Software-as-a-Service (SaaS) over the network. In this context, Software product lines (SPL) promote reusable application development for product families. On the other hand, business processes of most organizations are automated to support the advanced workflow management systems. In the context of the e-learning applications, it is perfectly suitable to manage the e-learning application by automating the e-learning process to effectively manage multiple tenants’ (institutions/teachers) ever changing requirements. As the available LMSs such as Moodle, Claroline and WebCT do not satisfy all the needs of different institutions/teachers, their demand is to develop their own systems. Cloud-based applications can be modeled as Business Process-as-a-Service (BPaas) families similar to the SPL products. As SPL development techniques rely on feature models to describe the commonality and variability of family member applications, such techniques can be used to model variability in BPaas. This paper proposes a generative multi-tenant e-learning process, that utilizes Business Process Feature Model (BPFM) approach which is a combination of SPL approach and BPM approach for modeling the reconfigurable e-learning process with multi-tenancy capabilities in cloud computing (BPaas).

The improvement of quality in business process models should be objective by applying measures. However, an evaluation of measurement results is not a simple task. It also requires the identification of relevant threshold values and indicators, which make it possible to distinguish between different levels of quality for BP models. In this context, a prototype named BPMoQualAssess (Business Process Model Quality Assess) is described. This prototype is including useful metrics defining the syntactic aspect of BP models also their threshold values. The obtained results are guidelines containing recommendations that aim at improving an input model by modifying it in order to obtain a higher level of quality.

The selection and the use of performance measurement have received considerable interest in recent years. In addition, it is crucial not only to track the process behavior and to derive the key performance indicators but also to understand all necessary concepts involved in the BP and incorporate domain knowledge of the field. The improvement of business processes is based on comprehensive measurement, data understanding, task design, and relevant result interpretation of organization's performance. In this context, in order to make relevant decisions and to provide a consistent understanding of the field, we present a new ontology based on a real business process to create semantic relationships between all terms. After that, we were based on data mining technique to extract the most important information from data measurement. An example of implementation of our proposed contribution as well as its validation on a real case study in the healthcare domain is presented.

During the last decades, flexibility has gained a strong presence, in a variety of disciplines, mainly in the BPM field. The real challenge for BPM consists in providing modeling paradigms and BPMSs with adequate information and features to deal with the often conflicting requirements of flexibility. In this setting, we focus on providing a guidance approach for enhancing business process flexibility. Our purpose is therefore to perceive which modeling paradigm(s) and/or business process management system(s) (BPMS(s)) are the most adequate to the specific organization needs in terms of flexibility. This approach was implemented in a plug-in named BPFlexGuide. To evaluate this approach, we have studied the emergency care (EC) process. Users interested in the EC process were guided to use the AristaFlow BPM suite BPMS. The results of this study would help designers to choose the best paradigms and BPMS that best fit their needs on flexibility.

Organizations need to continually improve and review their critical business processes. In addition, it is crucial not only to track the business process (BP) behavior and to derive key performance indicators (KPIs) but also to understand all necessary concepts and incorporate domain knowledge of the field. The purpose of this paper is to gain a deeper understanding of the interrelationships between all concepts and performance measurement raw data to extract their real meaning. In order to meet these challenges, first, we explore several qualitative and quantitative indicators for measuring the performance of BPs. Second, we develop a new ontology for the representation of these performance indicators. Then, we are based on data mining techniques to extract the most important information from data measurement and to discover all necessary relationships between indicators.

As e-learning becomes a basic need for several universities, a variety of Learning Management Systems (LMS) is proposed on the market. However, available LMSs do not satisfy all the needs of different institutions, which push them to develop their own systems. Since developing and maintaining new software are cost, time and effort consuming, and with the increasing demand on e-Learning systems, it becomes necessary to find an efficient solution that allows the fast development of systems and overcomes the afore-mentioned issues. We strongly believe that adopting a software product line approach in e-Learning domain can bring important benefits. We propose a general model for collaborative learning processes and we present the development process of an e-Learning software product line. Throughout the development process, we demonstrate how this approach allows us to satisfy the variable needs of universities/learners and benefit from the systematic large-scale reuse at the same time. In order to help organizations in providing similar services without the need to structure each of them separately, this paper presents how to support variability in learning process modeling.

Business process modeling is essential for improving and automating business processes and for comparing them. Furthermore, evaluating their quality must be based on a set of measures. In this context, our contribution in this paper consists in proposing a prototype for Process model assessment and applying it on a real business process model in the higher education. Our prototype is based on four main steps (modeling, measurement, evaluation and improvement). It takes as input a BPMN model for which some measures are applied to define its level of quality. After that, we compare the obtained measures with the threshold values presented in literature with the purpose of evaluating the considered BPMN model level of quality. Finally, based on guidelines applied to each exceeding measure, we help the designer to improve his/her model.

Business Process Management (BPM) is being increasingly adopted to streamline healthcare service delivery and management processes. Emergency departments (ED) are very complex. To operate effectively and create value, EDs must be flexible and have the ability to rapidly adapt to the highly variable needs of patients. In this paper, we provide models for a real-world scenario emergency care (EC) process, and elicit its flexibility needs using interviews with the EC process participants. We then use a guidance tool to help with the choice of the most suitable Business Process Management System (BPMS), regarding the flexibility needs elicited. We validate this choice by modelling the new flexibility enhanced EC process with one of the advised BPMSs, and analyse its support regarding the elicited flexibility needs.

Organizations that include several organizational units with similar business processes often suffer, with time, from the proliferation of processes variant models that significantly deviate from the original (to be followed) one. Take, for instance, student enrolment processes on distinct faculties of the same university, or healthcare processes of a National Health Service across distinct health centers. This can lead to poor global (process) management, since measuring and improving processes can be difficult with too many variants of the same business process. Related works analyse the generalisation and flexibility aspects of process models and related variants, but do not deal with the overall process model lifecycle, especially for this kind of organizations. This paper introduces a novel approach for the Governance and continuous improvement of process models for this kind of organizations. The approach is based on the general Business Process Management (BPM) cycle for process models, proposing concrete techniques for the phases of evaluation, classification and analysis between real and concrete variant models from each organizational unit. It includes the use of similarity metrics and flexibility in business processes, and the main output is a continuously improved template process model. This template foresees a common process part (best practice-based) including process elements collected from the process model variants verified in organizational units, and a flexible part, referring to possible (controlled) deviations that can be tolerated by the organization's headquarters. This approach enhances overall business process management and associated resources by enforcing uniform (good) behavior across similar organizational units. We present the results of our approach applied to a real-world case study of home-healthcare related business process models.