Our research activity focuses on the development of machine learning algorithms based on structured, semi-structured, and free-text recorded information. Our goal is to develop new data-driven machine learning algorithms that can effectively exploit and transform the hidden information inherent in exponentially growing data assets into valuable information.
We engage in the development of machine learning methods that fall into the realms of supervised, unsupervised, and reinforcement learning. A significant part of our research focuses on exploring the development and application potentials of neural networks and deep learning systems. The research outcomes offer both domain-specific and general-purpose solutions to address emerging questions and challenges.
Members of the laboratory:
Dr. Ágnes Vathy-Fogarassy, associate professor, head of the laboratory
Dr. Ágnes Stark-Werner, associate professor
Dr. Tibor Dulai, assistant professor
Dr. Gyula Ábrahám, research associate
Veronika Gombás, PhD student
Zsuzsanna Nagy, PhD student
János Kontos, PhD student
Tamás Miseta, PhD student
Péter Scsibrán, MSc student
Bendegúz Kósa, MSc student
Former member of the laboratory:
Dr. Dániel Leitold, assistant professor
Dr. Szabolcs Szekér
Key research outcomes:
- Development of a new method to prevent overfitting of neural networks (Correlation-Driven Stopping Criterion, CDSC)
- Development of neural network-based predictive and plausibility models in vehicle dynamics (based on real measurement data)
- Development of a neural network-based predictive optimization model for energy trading strategy for storage-based renewable power plants
- Development of a multi-level process mining methodology for analyzing disease-specific healthcare care processes
- Development of a machine learning method for classifying the effectiveness of chemotherapy treatment in lung cancer patients.
- Development of a model predicting heart failure complications in patients treated with anthracyclines and establishment of a risk-stratification model
- Identification of applicable drugs in the prevention of cardiotoxicity through the development of machine learning methods
- Development of a text mining methodology for structuring echocardiographic findings
Related publications:
T. Miseta, A. Fodor, Á. Vathy-Fogarassy (2024). Surpassing early stopping: A novel correlation-based stopping criterion for neural networks. Neurocomputing, 576, Paper: 127028
J. Kontos, B. Kránicz, Á. Vathy-Fogarassy (2023). Prediction for Future Yaw Rate Values of Vehicles Using Long Short-Term Memory Network. Sensors, 23(12), 5670.
S. Szekér, G. Fogarassy, Á. Vathy-Fogarassy (2023). A general text mining method to extract echocardiography measurement results from echocardiography documents. Artificial Intelligence in Medicine, 143, 102584.
T. Miseta, A. Fodor, Á. Vathy-Fogarassy (2022). Energy trading strategy for storage-based renewable power plants. Energy, 250, 123788.
J. Kontos, B. Kránicz, Á. Vathy-Fogarassy (2022). Neural Network-Based Prediction for Lateral Acceleration of Vehicles. In 2022 IEEE 2nd Conference on Information Technology and Data Science (CITDS) (pp. 153-158). IEEE.
Á. Vathy-Fogarassy, I. Vassányi, I. Kósa (2022). Multi-level process mining methodology for exploring disease-specific care processes. Journal of Biomedical Informatics, 125, 103979.
Z. Nagy, A. Werner-Stark (2020). A Multi-perspective Online Conformance Checking Technique. In: 2020 6th International Conference on Information Management (ICIM), IEEE, 172-176
Á. Vathy-Fogarassy, S. Szekér, B. Szolár, G. Fogarassy (2020). The Efficiency of Different Distance Metrics for Keyword-Based Search in Medical Documents: A Short Case Study. Studies in Health Technology and Informatics 271, pp. 232-239
D. Leitold, Á. Vathy-Fogarassy, J. Abonyi (2020). Network-Based Analysis of Dynamical Systems. Springer Briefs in Computer Science, p.110. Springer International Publishing
G. Fogarassy, Á. Vathy-Fogarassy, I. Kenessey, M. Kásler, T. Forster (2019). Risk prediction model for long-term heart failure incidence after epirubicin chemotherapy for breast cancer – A real-world data-based, nationwide classification analysis. International Journal of Cardiology, vol. 285, 47-52, 6 p.
G. Ábrahám, P. Auer, G. Dósa, T. Dulai, Á. Werner-Stark (2019). A Reinforcement Learning Motivated Algorithm for Process Optimization. Periodica Polytechnica-Civil Engineering, 63:4, 961-970
T. Miseta, Á. Vathy-Fogarassy (2019). The Effect of the Different Data Aggregation Methods and their Detail Levels to the Prediction of Bitcoin's Exchange Rate. In: Levente, Kovács; Carlos, M. Travieso-González (szerk.) Proceedings of IEEE International Work Conference on Bioinspired Intelligence IWOBI 2019, IEEE, pp. 145-152
Z. Nagy, Á. Werner-Stark, T. Dulai (2019). Using Process Mining in Real-Time to Reduce the Number of Faulty Products. In: Kamišalić Latifić, Aida; Podgorelec, Vili; Eder, Johann; Welzer, Tatjana (eds.) Advances in Databases and Information Systems: 23rd European Conference, ADBIS 2019, Springer, 89-104
S. Szekér, G. Fogarassy, K. Machalik, Á. Vathy-Fogarassy (2019). Application of Named Entity Recognition Methods to Extract Information from Echocardiography Reports. Studies in Health Technology and Informatics, vol. 260, 41-48
S. Szekér, Á. Vathy-Fogarassy (2018). The Effect of Latent Binary Variables on the Uncertainty of the Prediction of a Dichotomous Outcome Using Logistic Regression Based Propensity Score Matching. Studies in Health Technology and Informatics, vol. 248, 1-8
K. Tóth, K. Machalik, G. Fogarassy, Á. Vathy-Fogarassy (2017). Applicability of Process Mining in the Exploration of Healthcare Sequences. In: Szakál, Anikó (szerk.) IEEE 30th Jubilee Neumann Colloquium, 151-155
K. Tóth, I. Kósa, Á. Vathy-Fogarassy (2017). Frequent Treatment Sequence Mining from Medical Databases. Studies in Health Technology and Informatics, 2017, 236: 211-218
 |
Ágnes Vathy-Fogarassy (associate professor, head of Department of Computer Science and Systems Technology). She obtained the PhD degree at the Eötvös Lóránd University in 2009 in the field of Computer Science. She wrote her PhD dissertation in the topic of unsupervised learning (graph-based clustering algorithms). She habilitated in Computer Science in 2023. Her research interests include data-intensive predictive analytical methods, network analysis, deep learning, and applications of machine learning methods in healthcare. So far, she has co-authored 2 monographs and more than 100 articles, mainly related to the topic of the research laboratory. |