Semantic Clustering of Human-Animal Medical Corpuses

Michael Kane and colleagues from Yale University in the U.S. will create document clustering software incorporated into a web interface to enable clinical researchers to better search through the published literature on both human and veterinary medicine, to promote new discoveries for treating disease. Online biomedical literature and genetics databases carry large amounts of information on animal and human health. However, these two specialities diverge in the ways they are documented, making comparisons across species difficult with current online search engines, even in the context of a single disease such as Rift Valley fever, which infects both humans and animals. They will apply statistics and machine learning methods to enable large quantities of diverse data streams from the human and animal medical fields to be searched, thereby promoting new research directions for Rift Valley fever and other diseases.