MSc Lecture/Exercise Smart Process Analytics

+++ The first lecture is on: Thursday, 2018-10-25 room A213 (2pm s.t.). +++
This module consists of a lecture and corresponding tutorial.

Modul: 04WI2030

Lecturer

 

Information from KLIPS

Lecture

Topic

The lecture Smart Process Analytics (SPA) is on how we can support businesses by analyzing and interpreting data that is permanently generated by running software applications, smart devices, and/or general environmental services.

In particular, the lecture introduces how such an analysis can take place by using advanced concepts from the field of process mining. The topics that are covered by the course are as follows:

  • data collection and preprocessing
  • data analysis with process mining
  • deterministic process mining algorithms
  • probabilistic process mining algorithms
  • predictive process analytics
  • context-aware (=smart) process analytics

The exercise of the SPA course deepens the concepts introduced in the lecture by applying them in an extensive case study that has to be worked on in groups of 4-5 students. In the tutorial, the students are provided with process log data from a real-world business case, alongside with context information provided by sensors and/or online services. The task of the case study is to perform log data preprocessing, and, furthermore, smart (and, if applicable, predictive) process analytics with the concepts and tools introduced in the lecture. The results have to be presented and discussed in a final presentation session.

Important

The course SPA will be held in semi-blocked style. This means that from October to early December, we will mostly hold lectures BOTH AT LECTURE AND TUTORIAL TIMES. After that, we offer weekly tutorials that accompany the extensive case study work on process analytics, which has to be done in homework. The results of the case study must be presented at the end of the semester. In order to perform the blocked lecture most efficiently, we hold the lectures from 14h s.t. to 18h each Thursday.

Please note that you have to pass an inaugural short examination to get access to the course. We have to do this because in past semesters several students joined our courses who came with different expectations and different presumed knowledge, and who had problems to follow the courses later on. If you hold a Bachelor’s Degree from the School of Computer Science of the University of Koblenz-Landau (FB4), then you DO NOT NEED TO ATTEND THE INAUGURAL EXAM. The exam will be conducted on the first lecture date, which is Thursday, 2018-10-25 at 14h in Room A 213.

Please find the detailed schedule below.

Date Topic
Thu, 2018-10-25, 14:15h-15:45h Inaugural Examination
Thu, 2018-11-01 Holiday (All Saints)
Thu, 2018-11-08, 14h-18h Lecture: Introduction to Process Mining, exemplary Algorithms & Problems
Thu, 2018-11-15, 14h-16h Lecture: Log Preprocessing
Thu, 2018-11-15, 16h-18h Tutorial on Disco / Celonis
Thu, 2018-11-22, 14h-18h Lecture: Algorihms on Process Mining (deterministic & probabilistic)
Thu, 2018-11-29, 14h-18h Lecture: (smart) Algorihms on Predictive Process Mining
Thu, 2018-11-29, 14h-16h Lecture: (smart) Algorihms on Predictive Process Mining
from Thu, 2018-12-06
to Thu, 2018-12-20
Weekly tutorials (16:15-17:45h)
Thu, 2018-12-13, 14h-16h Presentation 1
2018-12-21 until 2019-01-19: Homework due to Holidays and Ski Seminar
from Thu, 2019-01-24 Still, weekly tutorials (16:15h-17:45h)
Thu, 2019-01-24, 14h-18h Intermediate presentation
Thu, 2019-02-07, 14h-18h Final presentation
Thu, 2019-02-17, 14h-16h 1st Exam
Thu, 2019-04-04, 14h-16h 2nd Exam
All regular lecture and exercise dates not named explicitly in this table are to be used for process mining and prediction in homework!

Language

The lecture will be held entirely in English to open it for non-German speakers. However, in case everyone speaks German, we will change to German.

The lecture and the exercise dates will not alternate regularly. I.e., the lecture will be held from October to December on both the lecture and exercise dates, and in December we will switch to the exercise.

Preconditions

It helps you (but it is not mandatory) if you know basics of business process modeling (ideally, you have already visited BPM). Students who do not hold a B.Sc. from the Faculty 4 (Fachbereich 4: Informatik) at the University of Koblenz need to pass a pre-test. Further Information will be published in the UniConnect Community.

How to get Credits

The course is worth 6 ECTS, and at the end, you will get one single combined grade. It will consist 50% of a written examination (90 minutes) on the topics of the lecture and 50% of the results of your case study work (this is because the exercise makes a considerable amount of work!).

To date, this course has not been added to any module guide (Modulhandbuch). Thus, the Link in KLIPS to the study programs has been made. To clarify Smart Process Analytics can be recognized in the programs as follows:

  • Master Wirtschaftsinformatik: Wahlpflicht Wirtschaftsinformatik
  • Master Informationsmanagement: Wahlpflicht Wirtschaftsinformatik
  • Master Web Science: elective subject
  • Master Informatik & Computervisualistik: You are welcome to join this course, but any recognition needs to be discussed with the responsible person for the regarding study program. 

Particular Topics and Timetable

All dates and topics can be found soon in UniConnect

Written Examinations

All dates and topics can be found soon in UniConnect

Literature and Material

If applicable, we will provide any relevant literature. All documents (if electronically available), slides and tools will be uploaded and made accessible on UniConnect. Inaugural reading is not necessary. We will introduce all relevant basics in the lecture.

Registration

Please register for the lecture on KLIPS so that we can estimate the number of participants. Furthermore, you need to register on UniConnect (after the first lecture) with your University of Koblenz e-mail address to get access to the literature and material pages.

Links: