The research mission of the Business Informatics group is to create and spread knowledge on how IT creates value for industry and society and to act as a catalyst for realizing IT’s potential. To achieve this mission, we are actively conducting research in the following key areas:
- Audit Analytics
- Process Analytics in Healthcare
- Behavioural Analytics
- Explainable AI
With audit analytics, we aim to assist the auditor in his work by improving current audit techniques and developing new techniques. Continuous auditing plays herein an important role: data analysis and process analysis methods can help in automatically performing some parts of the audit.
The main focus of our research group concerning audit analytics is process mining, which is a family of automated process analysis techniques. Other than Behavioural Analytics, we centre our research around the process mining type process discovery and conformance checking. On the one hand, the auditor can gain an understanding of the client’s business environment with process discovery. On the other hand, conformance checking enables the auditor to detect process deviations within the same business environment.
Our research aims to improve conformance checking techniques to make it fully adoptable for risk assessment in the context of a financial statement audit. For further questions about this topic or collaborations, you can contact prof. dr. Mieke Jans.
Process Analytics in Healthcare
Healthcare organisations have the ambition to provide high-quality services to their patients. At the same time, they are confronted with significant challenges such as tightening budgets, contrasted to increased care needs due to the ageing population. To face these challenges, these organisations are becoming increasingly aware of the need to manage their processes with the ambition to improve them.
In order to manage processes, such as the patient flow, healthcare organisations need to gain profound insights into these processes to, for instance, identify bottlenecks. These insights can be retrieved from data as increasing volumes of data are being recorded about the execution of healthcare processes. Traditionally, this data originated from the information system of the hospital. However, in the Internet of Things era, a variety of additional information sources such as real-time location systems and healthcare apps are available. This enables the retrieval of richer information to support hospitals in decision-making.
The research line process analytics in healthcare develops methods and techniques which support healthcare organisations to gather data-driven insights in their processes. Topics that are currently being developed include diagnostic analytics, data-driven capacity management and data quality assessment/improvement. Over the years, we have established collaborations with a multitude of healthcare organisations in Belgium and abroad. This ensures that we remain up-to-date with the sector’s specific challenges and, at the same, makes sure that research with societal impact is conducted.
Questions about this research line or seeing potential to collaborate? Please feel free to contact dr. Niels Martin.
Behavioural analytics concerns all data produced by a system describing events that occur over time within that system. The system can be defined in a very broad way: from user interaction on a website, data generated by a production site, logs originating from a server park, conversations of humans with chatbots, etc. The context of the problem to be tackled is therefore of the utmost importance.
The very same can be said about an event. The literal meaning of an event is a thing that happens or takes place, which as anything can be stored in databases. Where classic data would describe business objects, or the system as discussed earlier, themselves, event data describe the events concerning these business objects. This data structure not only allows the discovery of the current state of business objects, it also enables a rich analysis of its behaviour: by registering every click occurring on a website you not only know which pages were opened, additionally you can reconstruct the entire visiting experience for one specific user. This behavioural focus reflects the true value of event data.
Thus far, our research group’s main focus in the event data analytics field is process mining: discovering from data how business processes are truly executed in practice and how they can be improved. Another topic to be explored is the Internet of Things, a valuable event data generator.
Should you be interested in more information about this topic or our current work surrounding it, feel free to contact prof. dr. Benoît Depaire.
Explainable AI refers to the development of intelligent systems able to provide high quality, transparent solutions. These solutions provide an introspection mechanism to better understand how particular outcomes have been obtained. This feature plays a pivotal role in Business Intelligence: companies are less likely to implement automated intelligent solutions that neither stakeholders nor customers can understand. The acceptance of an intelligent model is not only a matter of trust, but also a matter of ethics and legality. Governmental agencies have moved to legislate the use of machine learning and increasingly require a level of insight and transparency into decision-making processes. As a result, the focus of machine learning is slightly shifting from pursuing more accurate models to improving the explanations behind those models.
While there is an increasing trend in developing post-hoc procedures to understand how black-box models operate, our research group has focused on developing inherently-interpretable intelligent systems such as Fuzzy Cognitive Maps. Our intelligent systems are not just capable of computing solutions with high levels of accuracy but of reasoning on the basis of expert knowledge, which often results in more realistic solutions. This means that our models are designed to reason together with human beings while being enhanced with available data records.
As a part of our research efforts, we have developed intelligent systems for classifying patterns, analysing time series, discovering segments and communities, reasoning with symbolic knowledge structures, among others. These solutions help stakeholders find relevant patterns in the data, which can be translated into more effective data-oriented decision-making processes.
Questions about this research line or seeing potential to collaborate? Please feel free to contact prof. dr. Koen Vanhoof.