PhD themes in EIT Digital Doctoral Training Centre with outstanding scholarship possibilities
Data analysis in financial systems
Optimization of internal processes in financial systems
The vast amount of structured and unstructured data that is generated by the internal processes of modern financial institutions requires efficient solutions for automated data preparation (collecting, cleaning), modelling, analysis and predictive modeling. The main task of the PhD process is to develop process- and data mining algorithms, which will enable financial systems to significantly optimize their internal processes on one hand, while maintaining the current information security posture in the financial environment on the other hand, e.g. the protection of personally identifiable information.
The most important research topics will be the following:
1) Collect information about the business processes which occur in the (target) financial system and formulate an initial set of hypotheses.
2) Identify business/analytical needs that will evolve in the near future as a consequence of technological progress, new data sources and changing business environment. Explore and implement current state-of-the-art solutions and on-going researches in the topic.
3) Create the first prototypes of models and automated data analysis and process optimization solutions. Test them on real-life business process datasets.
4) Improve the prototypes and deploy them in a production environment, or in an environment which is populated by realtime process data.
5) Evaluate the proper operation of the implemented solutions and make further developments based on empirical evidence, e.g. accuracy, satisfaction of key stakeholders in the financial institution, financial impact in the form of costs savings.
Industrial partner: OTP Bank
Academic/research supervisor: Lendák Imre
Application deadline: 30 May 2018
Customer profiling-based personalization in financial systems
The vast amount of structured and unstructured data that is generated in the retail banking sector requires efficient solutions for automated data preparation (collecting, cleaning) and predictive modeling. The main task is to create algorithms and solutions which effectively support decision making processes by exploring clients’ preferences and forecasting their future behavior. These goals will be achieved by creating and maintaining up-to-date customer profiles and developing algorithms and solutions which propose the optimal set of services to customers based on their profiles.
The most important research topics will be the following:
1) Collect information about the financial system, specify the most important pain points in behavior prediction and formulate an initial set of hypotheses.
2) Identifying business/analytical needs that will evolve in the near future as a consequence of technological progress, new data sources and changing business environment. Explore and implement current state-of-the-art solutions and on-going researches in the topic.
3) Create the first prototypes of models and automated data analysis and prediction solutions. Test them on real datasets.
4) Improve the prototypes and deploy them in a production environment, or in an environment which is populated by realtime financial data about customers.
5) Evaluate the proper operation of the implemented solutions and make further developments based on empirical evidence, e.g. accuracy, customer satisfaction, financial impact.
Industrial partner: OTP Bank
Academic/research supervisor: Lendák Imre