BioUSeR: a semantic-based tool for retrieving Life Science web resources driven by text-rich user requirements

Abstract

Background: OpenmetadataregistriesareafundamentaltoolforresearchersintheLifeSciencestryingtolocate resources. While most current registries assume that resources are annotated with well-structured metadata, evidence shows that most of the resource annotations simply consists of informal free text. This reality must be taken into account in order to develop effective techniques for resource discovery in Life Sciences. Results: BioUSeRisasemantic-basedtoolaimedatretrievingLifeSciencesresourcesdescribedinfreetext.The retrieval process is driven by the user requirements, which consist of a target task and a set of facets of interest, both expressed in free text. BioUSeR is able to effectively exploit the available textual descriptions to find relevant resources by using semantic-aware techniques. Conclusions: BioUSeRovercomesthelimitationsofthecurrentregistriesthanksto:(i)richspecificationofuser information needs, (ii) use of semantics to manage textual descriptions, (iii) retrieval and ranking of resources based on user requirements.