doc. Ing. Petr Hájek, Ph.D.

Doc. Ing. Petr Hájek, Ph.D., currently works as Associate Professor at the Institute of System Engineering and Informatics of the Faculty of Economics and Administration of the University of Pardubice. In 2006 he defended his dissertation thesis on the modelling of creditworthiness of municipalities using computational intelligence methods and in 2012 a habilitation thesis on the topic of credit rating modelling using soft computing methods. Doc. Hájek was a researcher of four projects of the Grant Agency of the Czech Republic focusing on the application of computational intelligence methods to support business decision-making. He was also a member of the research team of three other GA CR projects, one TA CR project and three projects commissioned by ministries.

Scientific interest:

• Fuzzy systems, neuro-fuzzy systems and evolutionary fuzzy systems
• Financial risk prediction using computational intelligence methods
• Deep learning neural networks for spam prediction
• Decision support systems under uncertainty

Scientific activities:

Doc. Hájek deals with methods of artificial and computational intelligence, especially fuzzy systems, neural networks, machine learning and systems integrating fuzzy systems, neural networks and evolutionary algorithms. He designed and applied a number of prediction models, including financial fraud prediction systems, economic performance prediction systems, social network and business platform spam prediction systems, and decision support systems in high-level uncertainty environments.

Prominent international cooperation:

• dr. Wojciech Froelich – University of Silesia in Katowice, Poland
• prof. Leonardo Vanneschi – Universidade de Lisboa, Portugal

Bibliometric indicators (Web of Science)

Number of papers 93, H-index 13, Total number of citations > 650

Citations per paper in WoS Categories 2000-2019: COMPUTER SCIENCE 6.61 (12.89), ECONOMICS 5.25 (6.17), BUSINESS 7.33 (5.58).

The most important projects:

• Modelling emotions in verbal and nonverbal managerial communication to predict corporate financial risk, 2019-2021, GAČR 19-15498S.
• Topic and sentiment analysis of multiple textual sources for corporate financial decision-making, 2016-2018, GAČR 16-19590S.
• The role of text information in corporate financial distress prediction models – country-specific and industry-specific approaches, 2013-2016, GAČR 13-10331S.

The most important publications:

• Hájek, P., Froelich, W.: Integrating TOPSIS with interval-valued intuitionistic fuzzy cognitive maps for effective group decision making. Information Sciences, 2019, vol. 485, s. 394-412.
• Papoušková, M., Hájek, P.: Two-stage consumer credit risk modelling using heterogeneous ensemble learning. Decision Support Systems, 2019, vol. 118, s. 33-45.
• Hájek, P.: Combining bag-of-words and sentiment features of annual reports to predict abnormal stock returns. Neural Computing and Applications, 2018, vol. 29, no. 7, s. 343-358.
• Hájek, P.: Predicting corporate investment/non-investment grade by using interval-valued fuzzy rule-based systems-A cross-region analysis. Applied Soft Computing, 2018, vol. 62, no. January, s. 73-85.
• Hájek, P., Henriques, R.: Mining corporate annual reports for intelligent detection of financial statement fraud – A comparative study of machine learning methods. Knowledge-Based Systems, 2017, vol. 128, s. 139-152.

Awards:

• Rector of the University of Pardubice Award 2019 for publication in a high AIS factor journal
• Rector of the University of Pardubice Award 2018 for a prestigious monograph entitled “Knowledge Spillovers in Regional Innovation Systems: A Case Study of CEE Region”
• 6× best paper award for best publication at international scientific conferences
• Applied Soft Computing Outstanding Reviewer Award for 2018