We offer 2 fully funded Postdoc/PhD positions for the OPJAK Trust project (https://cobra.cs.cas.cz/trust_job). Our group is part of the Department of Complex Systems, Institute of Computer Science, Czech Academy of Sciences.

The topic is focused on analysing modern and historical textual records through the lenses of graph-network analysis, statistics and machine learning. The topics include:

* Topic: Networks in Historical Scholarly/Philosophical Writings

Analysis of late medieval scholarly manuscripts (questions originating at the University of Prague around 1400) and presocratic textual fragments (mainly Heraclitus). Specific focus will be on (graph) clustering algorithms that should help to formally define relationships between semantically linked structures and thus offer insight into the structure of the textual source.

The next highly desirable part would be the development of image segmentation algorithms for recognising the overall structure of the yet never analysed formal parts of medieval handwritten manuscripts. Attempting full OCR (very hard in this case; see example on our web) and LLM interpretation of the distilled text would be a bonus.

Feel free to contact us informally for a more detailed description (Pavel Sanda, Institute of Computer Science, CAS, sanda@cs.cas.cz; Ota Pavlicek, Institute of Philosophy, CAS, will co-lead the medieval part).

* Topic: Information Flows in Scholarly & Scientific Networks

This topic aims to develop methodology to capture a realistic, interpretable, model of spreading of scholarly information in the scholarly social community. It shall include standard citation networks, but strive to extent to concept flow networks (detectable information influence beyond formal citations), and use graph theoretical, statistical and machine learning tools to provide further insights including identification of scholarly groups, informal knowledge propagation, extreme events including tipping points (paradigm shifts) and their early markers, influential agents and ideas.

The methodology will be piloted on mapping selected scholarly fields and historical periods in collaboration with domain experts in history and philosophy of science and particular scholarly/scientific fields.

Feel free to contact us informally for a more detailed description (Jaroslav Hlinka, Institute of Computer Science, CAS, hlinka@cs.cas.cz).

Conditions:

• Positions are available immediately.
• Applications will be reviewed on a rolling basis, until the positions are filled.
• Contract is for 12-36 months duration, with exceptions negotiable.
• Possibility of extension until end of project (31. 12. 2028), support for transition to tenure-track or external funding applications.
• This is a fixed term contract appointment, part time contract possible.
• Competitive salary based on qualifications and experience.
• Bonuses and travel funding for conferences and research stays depending on performance.
• No teaching duties.

Applicants should:

• Hold a PhD; with the exception of Research assistant and Graduate student positions.
• Have a strong background in the fields related to machine learning and artificial intelligence, experience with historical data analysis, text data analysis, network analysis, graph theory, multi-agent systems, and computer programming (the exact combination depends on the particular topic, see above).
• Be fluent in English.

We seek curious, self-motivated, hard-working, team-spirited researchers.

For description and other links see: https://cobra.cs.cas.cz/trust_job
For application see the formal position announcement: https://www.cs.cas.cz/job-offer/Scientific-positions-TRUST-Vidnerova-2025/en