Object-centric Predictive Monitoring has recently gained attention due to advances in machine learning and rise of Object-Centric Event Logs (OCELs), which comprehensively capture object interactions. This paper presents a modular framework supporting customizable pipelines for predictive analysis across diverse event logs. The framework comprises three core components: Preprocessing (preserving object relationships via graph structures), Graph Embedding Model, and Prediction Model. We experimentally evaluated various combinations of embeddings and predictors on three public OCELs. Results show that no single configuration consistently dominates. However, GAT and Graph Transformer models perform best for predicting remaining time and the number of events. Performance improves with larger embedding and subgraphs, particularly for neuralbased models. Finally, GAT delivered the most stable and highperforming results across all event logs in generalization tests.

In data science, a significant portion of time is dedicated to data preparation, a task often challenging for business users with limited technical skills. This research delves into using large language models (LLMs), particularly GPT (Generative Pre-trained Transformer), to streamline these data preparation processes. We study the usage of large language models in data cleaning, standardization, and transformation tasks. Our approach diverges from traditional methods by employing GPT as a direct transformation tool, utilizing its advanced reasoning capabilities through prompt engineering. This research aims to facilitate data transformation tasks, complicated by the diversity of data formats and inputs, for both technical and non-technical users, allowing them to accomplish these tasks through descriptive instructions rather than complex coding. We evaluate GPT’s effectiveness for frequent data transformation tasks and compare it against other data transformation tools on a benchmark. Our findings demonstrate the practical utility of GPT models in data transformation and propose prompt template guidelines for intricate data string transformations.

Event logs that are recorded by information systems provide a valuable starting point for the analysis of processes in various domains, reaching from healthcare, through logistics, to e-commerce. Specifically, behavioral patterns discovered from an event log enable operational insights, even in scenarios where process execution is rather unstructured and shows a large degree of variability. While such behavioral patterns capture frequently recurring episodes of a process’ behavior, they are not limited to sequential behavior but include notions of concurrency and exclusive choices. Existing algorithms to discover behavioral patterns are context-agnostic, though. They neglect the context in which patterns are observed, which severely limits the granularity at which behavioral regularities are identified. In this paper, we therefore present an approach to discover contextual behavioral patterns. Contextual patterns may be frequent solely in a certain partition of the event log, which enables fine-granular insights into the aspects that influence the conduct of a process. Moreover, we show how to analyze the discovered contextual behavioral patterns in terms of causal relations between context information and the patterns, as well as correlations between the patterns themselves. A complete analysis methodology leveraging all the tools presented in the paper and supplemented by interpretations guidelines is also provided. Finally, experiments with real-world event logs demonstrate the effectiveness of our techniques in obtaining fine-granular process insights.

Flexibility and adaptability have been praised and empirically validated as critical for an organization's performance. They are associated with the number of different behaviors that the organization exhibits, as well as with their level of sophistication, which in turn implies the number and the quality of the available knowledge and capabilities in place. In this work, we propose the creation of a bipartite network, involving organizations and behaviors, through an event log, to capture the prospects of both organizations and behaviors for process innovation. By using the Method of Reflections, we claim that the structure of that network can expose a set of sophistication metrics, that eventually suggest i) the potential of behaviors for better organizational performance; ii) the most reachable innovation paths; iii) an "order of significance" over the behaviors considering their improvement contribution power. We show the practical potential of our approach through a Proof of Concept based on a real dataset.

Data lakes have recently emerged as an alternative solution to costly traditional data warehouse solutions. To exploit data lakes, however, there is a need for means that assist users in combining and integrating data stored within a data lake. In this paper, we position ourselves in the recurrent context where a user has a local dataset that is not sufficient for processing the queries that are of interest to him/her. We show how data lakes, or more specifically the service lakes, since we are focusing on data providing services, can be leveraged to answer user queries, taking into account the quality of the services and respecting the (time and monetary) budget set by the user.

Emails represent a valuable source of information that can be harvested for understanding undocumented business processes of institutions. Towards this aim, a few researchers investigated the problem of extracting process oriented information from email logs to make benefit of the many available process mining techniques. In this work, we go further in this direction, by proposing a new method for mining process models from email logs that leverages unsupervised machine learning techniques. Moreover, our method allows to label emails with activity names, that can be used for activity recognition in new incoming emails. A use case illustrates the usefulness of the proposed solution.

International audience

In this paper, we propose an approach for approximate matching of OWL-S process model. We also propose a similarity measure that captures structural and semantic differences between two process models. To do so, we reduce the process matching to a graph matching problem and we adapt existing algorithms for this purpose.

With the growing importance of processes in current information systems and service oriented architectures, there is an increasing need for automatic techniques allowing to compare process models. Examples of such applications are numerous: delta analysis, version management, compatibility and replaceability analysis of business protocols, behavior based service discovery. When comparing two process models, first a mapping between their activities have to be found, identifying activities that are either equal or similar.

Finding useful services is a challenging and important task in several applications. Current approaches for services retrieval are mostly limited to the matching of their inputs/outputs. In this paper, we argue that in several applications (services having multiple and dependent operations and scientific workflows), the service discovery should be based on the specification of service behavior. The idea behind is to develop matching techniques that operate on behavior models and allow delivery of approximate matches and evaluation of semantic distance between these matches and the user requirements. To do so, we reduce the problem of behavioral matching to a graph matching problem and adapt existing algorithms for this purpose. To validate our approach, we developed a BPEL ranking platform that allows to find in a service repository, a set of service candidates satisfying user requirements, and then, to rank these candidates using a behavioral-based similarity measure.

The capability to easily find useful services (software applications, software components, scientific computations) becomes increasingly critical in several fields. Current approaches for services retrieval are mostly limited to the matching of their inputs/outputs possibly enhanced with some ontological knowledge. Recent works have demonstrated that this approach is not sufficient to discover relevant components. Motivated by these concerns, we have developed BeMatch platform for ranking web services based on behavior matchmaking. We developed matching techniques that operate on behavior models and allow delivery of partial matches and evaluation of semantic distance between these matches and user requirements. Consequently, even if a service satisfying exactly the user requirements does not exist, the most similar ones will be retrieved and proposed for reuse by extension or modification. We exemplify our approach for behavioral services matchmaking by describing two demonstration scenarios for matchmaking BPEL and WSCL protocols, respectively. A demo scenario is also described concerning the tool for evaluating the effectiveness of the behavioral matchmaking method.

The capability to easily find useful services (software applications, software components, scientific computations) becomes increasingly critical in several fields. Current approaches for services retrieval are mostly limited to the matching of their inputs/outputs. Recent works have demonstrated that this approach is not sufficient to discover relevant components. In this paper we argue that, in many situations, the service discovery should be based on the specification of service behavior (in particular, the conversation protocol). The idea behind is to develop matching techniques that operate on behavior models and allow delivery of partial matches and evaluation of semantic distance between these matches and the user requirements. Consequently, even if a service satisfying exactly the user requirements does not exist, the most similar ones will be retrieved and proposed for reuse by extension or modification. To do so, we reduce the problem of behavioral matching to a graph matching problem and we adapt existing algorithms for this purpose. A prototype is presented (available as a Web service) which takes as input two conversation protocols and evaluates the semantic distance between them; the prototype provides also the script of edit operations that can be used to alter the first model to render it identical with the second one

In this paper we present a process management technology for the coordination of creative and large scale distributed processes. Our approach is the result of usage analysis in domains like Software Development, Architecture/Engineering/Construction, and e-Learning processes. The basic conclusions of these experiments are the following: (1) cooperative processes are described in the same way as production processes, but these descriptions are interpreted in a different way depending on the nature of the process, (2) the interpretation of process description depends mainly on the required flexibility of control flow and of data flow, and on the relationship between them, (3) the management of intermediate results is a central feature for supporting the cooperation inherent to these processes. COO-flow is a process technology that results from these studies. It is based on two complementing contributions: anticipation that allows succeeding activities to cooperate, and COO-transactions that allows parallel activities to cooperate. This paper introduces COO-flow characteristics, gives a (partial) formalization and briefly discusses its Web implementation.