There will be four sets of lectures by the following distinguished lecturers:
- Wil van der Aalst -- Process mining
- Stefan Edelkamp -- Graph Search Engineering
- Charles Elkan -- Machine learning and bioinformatics
- Kurt Jensen -- Coloured Petri Nets
Process Mining: Beyond Business Intelligence
This tutorial aims to provide an overview of process mining techniques and, using many real-life examples, it will be shown how particular techniques can be applied and what kind of insights such analyses provide.
More and more information about processes is recorded in the form of event logs. Equipment ranging from embedded systems to enterprise information systems are logging the behaviors that take place. This data explosion allows for the analysis of reality and the construction of models that reflect what actually happened. This can be used to diagnose and improve processes in a variety of domains. Especially business processes involving human actors are interesting to diagnose because these processes are not controlled by software and there may be a gap between what people think that happens and what really happens. Process mining provides a versatile and extendible way to analyze such processes. using process mining techniques it is possible to extract different types of models from event logs, e.g., the construction of a process and organizational models. Moreover, other techniques support the conversion and analysis of models. Using conformance checking techniques models can also be compared with reality and existing models can be enhanced with additional information, e.g., indicating bottlenecks in a process. Many vendors claim to offer support for Business Intelligence (BI). Unfortunately, these BI tools are not intelligent at all. Moreover, these tools require input data of a particular type and a predefined model. Process mining overcomes these limitations and makes it possible to extract new knowledge from information systems in a truly intelligent way. Process mining addresses the problem that most organizations have very limited information about what is actually happening in their organization. In practice, there is often a significant gap between what is prescribed or supposed to happen, and what actually happens. Only a concise assessment of the organizational reality, which process mining strives to deliver, can help in verifying process models, and ultimately be used in a process redesign effort.
The tutorial is intended for both researchers and practitioners in the area of business process management. It is assumed that people have a basic understanding of business process management and are familiar with the basics of process modeling using languages such as BPMN, EPC, or similar. It aims at an audience that is interested in the analysis of processes as they actually happen in reality.
Graph Search Engineering
Graph Search algorithms and their variants play an important role in many branches of computer science. All use duplicate detection in order to recognize when the same node is reached via alternative paths in a graph. This traditionally involves storing already-explored nodes in random-access memory (RAM) and checking newly-generated nodes against the stored nodes. However, the limited size of RAM creates a memory bottleneck that severely limits the range of problems that can be solved with this approach. Although many clever techniques have been developed for searching with limited RAM, all eventually are limited in terms of scalability, and many practical graph-search problems are too large to be solved using any of these techniques.
Over the past few years, several researchers have show that the scalability of graph-search algorithms can be dramatically improved by using external memory, such as disk, to store generated nodes for use in duplicate detection. However, this requires very different search strategies to overcome the six orders-of-magnitude difference in random-access speed between RAM and disk. We consider recent work on external-memory graph search, including duplicate-detection strategies (delayed, hash-based, and structured); integration of these strategies in external-memory versions of breadth-first search, breadth-first heuristic, and frontier search; the inclusion of perfect hash function, as well as combining parallel and disk-based search; external-memory pattern-database heuristics; and applications of external-memory search to AI planning, automated verification, and other search problems. Implicit graph search that is in the scope of the lectures includes deterministic and non-deterministic models, as well as game-theoretical and probabilistic models.
Moreover, the lectures are specifically concerned with algorithm designs for implicit graph search on modern personal computer architectures, e.g. subject to several processing units on the graphic card, and hierarchical memory including solid state disks. Applications areas for new algorithm designs that exploit modern hardware are found in the model checking community, but also in AI planning and game-playing.
Introduction to Coloured Petri Nets
This course will focus on Coloured Petri Nets (CPN) which is a graphical modelling language used for many different kinds of systems. CPN allows system designers to build models that can be executed and analysed by a computer tool. Simulation of CPN models makes it possible to conduct a detailed investigation of the system behaviour, and reason about performance properties (such as delays and throughput). State space analysis makes it possible to verify functional properties of the system (such as absence of deadlocks).
The course will illustrate how coloured Petri Nets can be used to model and analyse different kinds of systems ranging from technical systems (such as communication protocols) to workflow systems (such as the work processes on a hospital).