Projects > Rosetta

Rosetta

Rosetta is an information retrieval system for scientific literature. Rosetta makes it easy for people to find the information for which they are looking even when using short, imprecise queries. Rosetta indexes Computer Science research articles based on the way they have been described when cited in other documents. Citations are an exceptional data source to use in indexing, because they label only the important topics a document discusses. Furthermore, in doing so they describe which papers other people have found most useful on those topics. We use this indexing technique to present users of Rosetta with a powerful user interface supported by three information-gathering tools. The first is a fast and accurate retrieval tool for subject searching. The second is an automatic information directory generator that presents users with a navigable overview of the information space surrounding a given query. The third is an automatic information type identifier and retrieval tool that suggests the type of information available (introduction, overview, evaluation metric etc.) to a user for a given query and also allows users to search for a particular type of information. The publications on this work present overviews of the three tools that comprise Rosetta’s interface and describe studies that demonstrate the benefits of this information retrieval technology.

Project Papers

  • Automatically Indexing Documents: Content vs. Reference
     Download
  • Using citations to facilitate precise indexing and automatic index creation in collections of research papers
     Download
  • Guiding People to Information: Providing an Interface to a Digital Library Using Reference as a Basis for Indexing
     Download
  • Constructing indices from citations in collections of research papers
     Download