To address the challenging task of extracting information on lifestyle from text, we have developed a prototype lifestyle-factor ontology. You will train state-of-the-art deep learning-based language models to identify lifestyle descriptors, such as diet, from one of the large biomedical text collections comprising millions of full-text articles. Subsequently you will use it in combination with existing pretrained models and vocabularies to resolve cases of ambiguity, extract associations between diseases and lifestyle factors, and construct an open publicly available knowledge graph from the results.
In collaboration with other members of the groups, you will also use cross-lingual representation learning to identify lifestyle factors in text from Danish electronic health records. Furthermore, you will use the literature-based knowledge graph to interpret patient-level data on the progression from healthy to sick from electronic health records and registry data.
Projected start date: Second quarter of 2021
Application deadline: March 15th, 2021.
For more details please see https://jobportal.ku.dk/videnskabelige-stillinger/?show=153544
_______________________________________________A postdoc position is available jointly between the Cellular Network Biology (Lars Juhl Jensen) and the Translational Disease Systems Biology (Søren Brunak) groups at The Novo Nordisk Foundation Center for Protein Research (CPR – https://www.cpr.ku.dk). The Disease Systems Biology program consists of three research groups covering many systems level aspects of biology and medicine, including the integration of molecular-level data and healthcare data, including biomedical texts.To address the challenging task of extracting information on lifestyle from text, we have developed a prototype lifestyle-factor ontology. You will train state-of-the-art deep learning-based language models to identify lifestyle descriptors, such as diet, from one of the large biomedical text collections comprising millions of full-text articles. Subsequently you will use it in combination with existing pretrained models and vocabularies to resolve cases of ambiguity, extract associations between diseases and lifestyle factors, and construct an open publicly available knowledge graph from the results.
In collaboration with other members of the groups, you will also use cross-lingual representation learning to identify lifestyle factors in text from Danish electronic health records. Furthermore, you will use the literature-based knowledge graph to interpret patient-level data on the progression from healthy to sick from electronic health records and registry data.
Projected start date: Second quarter of 2021
Application deadline: March 15th, 2021.For more details please see https://jobportal.ku.dk/videnskabelige-stillinger/?show=153544
--Lars Juhl Jensen
Professor, Group Leader in Disease Systems Biology
NNF Center for Protein Research
Faculty of Health Sciences
University of Copenhagen, Denmark
http://jensenlab.org
SocBiN mailing list
Soc...@lists.su.se
https://lists.su.se/mailman/listinfo/socbin-at-sbc.su.se