Online Spatial Prediction Model for Citizens’ Public Space Complaints in Eindhoven

  Smart cities approach does not only emphasize the implementations of new technologies in a city but also highlights the importance of using new technologies for enabling citizens’ engagement in urban planning processes. In that regard, ICTs play a vital role in (i) supporting citizens to report their complaints related to the public spaces (i.e. Read More …

Develop a Behavioral Event Data Query Language

Query languages are essential for exploring, working with data and directly answering questions from data. SQL is the prime example for answering questions on relational data. Behavioral data is recorded in the form of events with timestamps. Various techniques such as Process Mining use the data in the form of event logs to aggregate and Read More …

Mercedes-Benz Customer Assistance Center in Maastricht

Mercedes-Benz are well recognized as industry leaders in luxury service and high quality products, pushing the meaning of automobile excellence to new boundaries. You can trace the timeline of Mercedes-Benz all the way back to 1885 when Karl Benz invented the first automobile – cementing Mercedes-Benz a place in history. We know what you’re thinking, Read More …

Emergency and Operating rooms at HagaZiekenhuis

Two of the most important parts of a hospital are the Emergency room and the Operating room. HagaZiekenhuis deals with tens of thousands of patients every year. For each patient, everything is recorded in the Electronic Patient Record (Chipsoft Hix). Exploring the data recorded by both areas of the hospital opens the possibilities to understand Read More …

Happy nurses

The biggest problem that Dutch hospitals face today is the lack of nursing staff. Therefore, hospital beds are closed to maintain safety levels for the already admitted patients. However, patients on the waiting list or those already planned for elective care experience longer waiting times. And sometimes emergency care beds have to be closed causing Read More …

Real-Time Prediction of Traveler Flow within Digital Stations

In the last decennia the pressure on different types of mobility have severely increased in the Netherlands. Therefor the need for availability and reliability has increased. Siemens Mobility supplies solutions in the Netherlands that contribute to the accessibility and quality of life in this regard. With the help of different technologies, data is being unlocked Read More …

Improving Traffic Flow Prediction in Urban Areas by Incorporating a Real-Time Outlier Detection Model

In the last decennia the pressure on different types of mobility have severely increased in the Netherlands. Therefor the need for availability and reliability has increased. Siemens Mobility supplies solutions in this regard that contribute to everyday accessibility and quality of life. With the help of different technologies, data is being unlocked through which the operation Read More …

Differential-private Process Mining (Multiple Assignments)

Within the BPR4GDPR EU project, we are researching (among others) methods that enable a privacy-aware utilization of sensitive individual information. Several anonymization techniques are not enough to completely keep the process discovery completely privacy aware (e.g. the existence of rare diseases can still be revealed from an anonymized log file). Adding exactly “the correct amount” of Read More …

Log-based vs. Model-based Concept Drift Detection

StrProMCDD is a recently published work that detects concept drifts in event streams (see the figure below). StrProMCDD uses several model-based distance measures to detect these deviations using an adaptive window concept. In this assignment, we would like to compare the performance of this model-based approach with log-based stream clustering approaches that try to detect drifts in Read More …

Process Discovery using Generative Adversarial Neural Networks

Process Discovery is an unsupervised learning problem with the task of discovering a graph-based model from sequences (or graphs) of event data that describes the data best. Generative Adversarial Neural Networks (GANNs) are a type of neural networks used to learn structures in an unsupervised fashion. The objective of this project is to explore the Read More …

Process Mining on Event Graph Databases (multiple projects)

Process mining assumes event data to be stored in an event log, which is technically either a relational table (attributes as columns) or a stream of events (attribute value pairs). Recently, we developed a new technique to store event data in a Graph database such as Neo4j. This allows to do process mining over various Read More …

Mining processes, social networks, and queues (multiple projects)

A recent visual analytics technique called the “Performance Spectrum” https://github.com/processmining-in-logistics/psm allows us to gain more fine-grained insights into performance behavior and changes over time. A TU/e Master student showed that it is possible to mine synchronization of cases from the performance spectrum data showing that the behavior of a case depends on the mechanisms and Read More …

Process Mining with Textual Data

In many application domains, a process execution is captured using natural language. Think of medical records, customer complaints, legal records… The same holds for process models: they can be captured as text for medical guidelines, user manuals, legal regulations are typical examples of such cases. Such data forms a new challenge for the process mining Read More …

Finding Patterns in Evolving Graphs

The analysis of the temporal evolution of dynamic graphs like social networks is a key challenge for understanding complex processes hidden in graph structured data. Graph evolution rules capture such processes on the level of small subgraphs by describing frequently occurring structural changes within a network. Existing rule discovery methods make restrictive assumptions on the Read More …

Efficient unsupervised event context detection

for event log clustering, outlier detection, and pre-processing. We recently developed a technique to detect the context of events from an event log in an efficient way through sub-graph matching. This allows to identify events and parts of event logs which are similar or different to each other, allowing to cluster traces, detect outliers, and Read More …

Smart event log pre-processing

The quality of process mining results highly depends on the quality of the input data where noise, infrequent behaviors, log incompleteness or many different variants undercut the assumptions of process discovery algorithms, and lead to low-quality results. ProM provides numerous event log pre-processing and filtering options, but they require expert knowledge to understand when which Read More …

Log Data Anonymization

In the context of process mining, we are often confronted with companies willing to share their data if we can sufficiently anonymize this. However, to date, there are no well-defined plugins to do such anonymizations. Therefore, we are looking for a Master student that is willing to help us with this. Part of the project Read More …

Adding heuristics to the Block Layout

The Block Layout can be used to create a layout for a process graph. For this, it uses well-known Petri-net-based reduction rules to reduce the entire net into a single place. For nicely structured process graphs, this layout works quite well, but for more complex structured graphs, the resulting layout needs to be improved. Either Read More …

N-out-of-M patterns in alignments

Aligning structured process models to event logs is a far from trivial task. In complex modelling languages, inclusive OR-split/join patterns play an important role and they are known to be notoriously difficult to align to event logs due to their large state-spaces. The known Petri net translations of OR-joins rely either on token coloring or Read More …

Generating non block-structured models and corresponding logs

For experimenting with process discovery and Petri nets, scientists often rely on experiments with artificial models and logs. More often than not, these models are block structured as it is easy to generate such models by simply building a random process tree and translating that into a Petri net. However, Petri nets allow for more Read More …

Petri net reduction rules for replay

Replaying event logs on Petri nets, either through token-replay or using alignments, is a complex task. Especially when models become larger and have more labels, the size of the models becomes a problem. In Petri net theory, many reduction rules exist for reducing Petri nets while retaining, for example, soundness of the model. Can we Read More …