It can be possible that payment terminals that are installed and maintained by Worldline generate malfunctions or do irregular reboots throughout their lifetime. If a malfunction occurs, data is sent from the terminal to the servers of Worldline. This data describes the event that occurred with various attributes. The data that is sent from the terminals to the servers of Worldline provides valuable information for Worldline, which is not yet extracted. The key point of this project is therefore to extract knowledge from this data to improve the reliability of terminals in several ways.

The project was split into three agile phases. When a phase was completed, the results and findings of the previous phase are the foundation of the following phase. In the first phase, the data and the system environment were understood. Additionally, data were visualized in a user-friendly and understandable way. In a second phase, the results were used to define data mining methods and tools to continue to work with (MS Power BI, Python Pandas). The last phase consisted of implementing the selected data mining methods.

As previously mentioned, the logs that are sent from the terminals to the Worldline Servers provide valuable information for Worldline. With the help of the analysis, the client is able to start improving terminals (hardware), terminal software and processes. This will increase the quality of the services Worldline offers. 1. Glossary and priority of events 2. Identification of at least three data mining methods 3. Visualization of data identification 4. Knowledge for the client created based on the analysis of the provided data