A Knowledge-Poor Approach to BioCreative V DNER and CID Tasks

The report describes our participation in the BioCreative V track #3, both in Disease Named Entity Recognition and Normalization (DNER) and in Chemical-induced diseases relation extraction (CID). For both tasks, we have adopted a general-purpose approach based on machine learning techniques integrated with a limited number of domain specific knowledge resources and using freely available tools for preprocessing data. Crucially, the system only uses the data sets provided by the organizers. After comparing different configurations, the one giving the best compromise between effectiveness and efficiency has been chosen. We report the results of the experiments performed during the development phase for comparing different configurations. The results of the official submission are in line with those on the development set.