@inproceedings{PhilippMoralesGeorgiBeyereretal.2019, author = {Philipp, Patrick and Morales Georgi, Rafael X. and Beyerer, J{\"u}rgen and Robert, Sebastian and Beyerer, J{\"u}rgen}, title = {Analysis of Control Flow Graphs Using Graph Convolutional Neural Networks}, series = {2019 6th International Conference on Soft Computing \& Machine Intelligence (ISCMI)}, booktitle = {2019 6th International Conference on Soft Computing \& Machine Intelligence (ISCMI)}, pages = {73 -- 77}, year = {2019}, abstract = {With the digital transformation of companies, ever larger amounts of data are generated and available for analysis. Process mining techniques can be used to extract and analyze process models from these data. Related techniques have quickly developed into an important field with constantly increasing investments in recent years. Thus, the automated analysis of processes has gained an important role in many companies. In this context, graphs have been shown to be an intuitive representation of how the gathered processes are carried out using the aforementioned techniques. For the analysis of these so-called control flow graphs, we investigate the use of convolution neural networks, which are specially designed for graphs: graph convolution networks (GCNs). In our contribution, GCNs are used to perform a regression task based on individual control flows of a process in which farmers apply for specific governmental payments. The approach achieved promising results on this publicly available data set.}, language = {en} } @techreport{Seidlmeier2024, author = {Seidlmeier, Heinrich}, title = {Hidden networks in business processes - theoretical considerations and experimental verification}, doi = {10.13140/RG.2.2.24967.00161}, pages = {22}, year = {2024}, abstract = {The discipline of process mining was the first to demonstrate the construction of social networks from business processes. Collaboration in the process flow creates relationships between the participants. The objective of this article is to further develop this basic idea of "hidden process-induced networks" (in short: process networks) on the basis of the findings of social network research and to make a new contribution to the management and design of business processes. A contingency network theory of processes is presented for the first time as the methodological foundation for the article. In essence, it is demonstrated how the characteristics of a process network (contingency factors) can be utilized as design variables for achieving desirable process improvements. This is followed by an experimental verification and confirmation of this theory of business process design. Finally, use cases demonstrate the practical applicability of the findings. Based on experimentally obtained data, regression analyses show that there are strong correlations between certain network measures (quantitative network characteristics) and the process measures processing time and cycle time. Processes therefore not only form networks. Networks can also support the optimization of processes. Process networks can be regarded as novel, theoretically sound and empirically verified explanatory and design models for business process management.}, language = {en} }