2021/12/20 by Draheim, Christina, Wawrzinek, Janus, Wulle, Stefan +1
#Discovery System #Fachinformationsdienste #Medicine and health #Pharmazie #PubPharm #discovery system #machine learning #maschinelles Lernen #pharmacy #specialised information services
paper · doi:10.3205/mbi000509
In the Specialised Information Service Pharmacy, the cooperation between the University Library and the Institute for Information Systems at Technische Universität Braunschweig results in innovative data retrieval services. Based on basic information technology research, prototypes are developed and evaluated by focus groups of the pharmaceutical research community. Upon successful prototype optimisation, services are implemented in the freely available search platform PubPharm ([link:https://www.pubpharm.de*https://www.pubpharm.de]). Recent artificial intelligence models (in detail neural networks) can automatically determine associations between pharmaceutical entities (i.e. clearly definable concepts such as active ingredients, diseases and genes) based on research literature. This research activity at the Institute for Information Systems resulted already in a web service for PubPharm: listing contextualised substances, diseases or genes when searching for active substances or diseases/symptoms. The Drug-Disease-Networks complement this suggestion function and visualise the partly complex entity relationships supporting their interpretation. Links to external data sources, e.g., Comparative Toxicogenomics Database and DrugBank, provide further information.