About the Role
Scientist (m/f/d) in the field of GeoAI with a focus on multilingual geo-parsing (Vacancy ID 6423)
Place of work: Jena, Germany
Starting date: as soon as possible
Career level: Graduates, experienced professionals
Type of employment: Part time, full-time
Duration of contract: up to 3 years
Remuneration: In accordance with the Collective Agreement for the Public Sector – Federal Government (TVöD-Bund).
The Institute of Data Science in Jena supports all DLR application areas with data science concepts, methods and tools. The Multimodal and Spatial Information Retrieval (DMO-MIR) group in the Data Acquisition and Mobilization department researches practical methods and modular software for the collection, processing, filtering, geolocation and content evaluation of unstructured data.
Your tasks
Contribute to the DFG project GeoMMCorpus+Eval
Familiarise yourself with the state of research in GeoAI, e.g. multilingual geo-parsing, creation and annotation of geographical text corpora, and LLM-supported annotation and quality assurance
Design scientific methods and workflows for extracting, processing, selecting and annotating large, heterogeneous, multilingual text datasets
Develop and apply data science methods to extract, clean, filter and structure large multilingual corpora (websites, news, social media, academic publications, historical documents)
Plan experimental studies on LLM-assisted annotation and iterative human-in-the-loop quality assurance
Design and apply quality metrics to evaluate results
Qualifications
Master's / university diploma in Computer Science, Data Science, Geoinformatics, Computational Linguistics or a related field
In-depth knowledge and practical experience in NLP, information extraction and machine learning
Experience with large language models (prompting, fine-tuning)
Python programming and version control (Git)
Experience processing and analysing large, heterogeneous text datasets
Ability to plan, carry out and evaluate scientific experiments independently
Very good written and spoken English
Knowledge of geoinformation processing, GeoAI or geographic information extraction from texts is an advantage
Contact: Dr. Friederike Klan, Tel. +49 3641 30960 555