PresseBox
Press release BoxID: 145833 (Transinsight GmbH)
  • Transinsight GmbH
  • Tatzberg 47-51
  • 01307 Dresden
  • http://www.transinsight.com
  • Contact person
  • +49 (351) 796 57 80

Transinsight’s GoPubMed with Social Networking Features for Biomedical Experts

GoPubMed now allows users to identify experts in the biomedical field and gain important information on recent research topics by viewing their networks

(PresseBox) (Dresden, Germany, ) Biomedical research happens in networks of researchers. Social networking web sites like FaceBook, LinkedIn and Xing use personal networks to establish contacts. On these sites, however, connections must be defined by the users themselves. For the first time, GoPubMed (http://www.GoPubMed.org) now completely and automatically extracts collaboration networks from millions of biomedical science publications. For each concept in the selected semantic background knowledge, GoPubMed’s "Hot-Topic-View" shows the collaboration network between top authors in this field of research. Collaboration networks can now be experienced and visualized. GoPubMed also now allows these networks to be searched for possible experts and collaboration partners, a feature which leads to tremendous time saving when searching for appropriate experts. This feature is especially important in a specialized scientific world where it is becoming more and more vital to set up temporary teams of highly specialized experts.

"Some author names like Lee S., Smith J. and Müller C. appear over 20,000 times! We have solved the technical challenge of disambiguating the authors into single individuals with our semantic search technology, which in a way functions like the network of the brain", says Prof. Dr. Michael Schroeder, CSO and co-founder of Transinsight. If two articles share the same author, GoPubMed evaluates their similar properties. The system hereby takes into account that the author of each paper often publishes about similar research topics, with the same co-authors and in the same journals. The research topics are thereby connected to the concepts of the semantic network in the background. The more concepts two articles have in common and the shorter the semantic distance in the network is, the more likely it is that the articles were written by the same person. This approach leads to impressive accuracy. If at any point the system is not correct, it can be corrected by the users.

"GoPubMed is an essential step in significantly easing the finding of complexly networked information", according to Prof. Dr. Michael Brand, Director of the BioInnovation Center in Dresden. "The semantic approach is unparalleled worldwide, and I’m excited that such a development, which would be expected to come from Palo Alto’s Stanford University in the Silicon Valley in California, today comes instead from Dresden," says Brand.

See also http://www.transinsight.com/interview