Our Research Projects
=====================

### ActiTraC

ActiTraC is a novel trace clustering approach based on active learning which significantly outperforms other clustering techniques when applied on complex, large event logs. [→ Read More](/actitrac)

### ActiTraC-MO

ActiTraC-MO is an adapted version of ActiTraC that constructs traces clusters in parallel rather than sequential, allowing for a multi-objective approach to trace clustering. [→ Read More](/multiobjective)

### AGNEsMiner

AGNEsMiner is a declarative machine learning technique that represents the problem of process discovery as a multi-relational classification problem on event logs supplemented with Artificially Generated Negative Events (AGNEs). [→ Read More](/agnesminer)

### CoBeFra

CoBeFra is a comprehensive benchmarking suite to set up large scale conformance checking experiments and is used and loved by many researchers in the field. [→ Read More](/cobefra)


### Context-aware Petri nets

Context-aware Petri nets (CAPN) enable the straightforward use of ontologies in Petri nets. [→ Read More](/capn)

### Declare Execution Environment

This tool allows users to enact Declare models that are annotated with hidden dependency information. [→ Read More](/declareexecutionenvironment)

### Determining the Number of Trace Clusters: a Stability-based Approach

Determining the appropriate number of trace clusters in an event log based on the stability of the clustering. [→ Read More](/clusterstability)

### Expert Knowledge in Trace Clustering

ActiTraC-SemiSupervised is an adapted version of ActiTraC that incorporates expert knowledge into its clustering approach. [→ Read More](/expertdriventraceclustering)

### Fodina

Fodina is a cutting-edge robust, flexible, and scalable heuristic dependency-based process discovery technique. [→ Read More](/fodina)

### Fusion Miner

Fusion Miner is the first process discovery approach for mixed-paradigm process models with intertwined state spaces. [→ Read More](/fusionminer)

### FusionMINERFul

FusionMINERFul is a fast mixed-paradigm process mining algorithm with self-learning flexibility analysis and model checker. [→ Read More](/fusionminerful)

### Location-Aware Process Modeling and Execution

Integrating location-aware aspects with process flow concerns using Coloured Petri nets. [→ Read More](/locationaware)

### Negative Event Based Conformance Checking

Blazingly fast and precise conformance checking using negative events. For recall, precision, and generalization assessments. [→ Read More](/neconformance)

### Petri net Event Log Generator

An easy-to-use Petri net driven event log generator with timing, resource, and flow options. Great to set up experiments with synthetic models. [→ Read More](/loggenerator)

### P-MInD

P-MInD is a framework for discovering holistic decision models from event logs. [→ Read More](/PMInD)

### Representation Learning for Business Processes

Embeddings applied in the world of business processes: act2vec, trace2vec, log2vec and model2vec. [→ Read More](/replearn)

### Real-time Decomposed Conformance Checking

Step one: decompose a model. Step two: apply negative event based conformance checking. Step three: visualize and monitor conformance issues and deviations in real time. [→ Read More](/rtconformance)

### Rozinat's Conformance Framework Port for ProM 6

Dr. Anne Rozinat's conformance checker framework remains one of the most well-known and seminal conformance checking frameworks in the process mining community. Sadly, its implementation—a ProM 5 plugin—has never been ported to ProM 6. Until now. [→ Read More](/rozinatconformance)

### Rozinat's Decision Mining Framework Port for ProM 6

Dr. Anne Rozinat's decision mining framework port for ProM 6. [→ Read More](/rozinatdecmin)

### SECPI

Explaining why process instances are put in a particular cluster based on machine learning on features generated from control flow aspects. [→ Read More](/svmexplainer)

### A Stability-assessment Framework for Process Discovery Techniques

Assessing the stability of process discovery techniques based on log-perturbation and process model similarity. [→ Read More](/PDStability)