Semantic Survey Methodology

Despite significant advances in technology, the way how research is done and especially communicated has not changed much. We have the vision, that ultimately researchers will work on a common knowledge base comprising comprehensive descriptions of their research, thus making research contributions transparent and comparable. In this article, we describe how surveys on research fields can be performed in a semantic way resulting in a knowledge graph describing the individual research problems, approaches, implementations and evaluations in a structured, comparable way. We illustrate our methodology with the example of ...

ToDo

• select a narrow research field with many comparable approaches as an example - e.g. question answering, link discovery, named entity recognition, relationship extraction,
• select 10-20 papers in this field
• create pages for all papers in OpenResearch comprising key semantic information about the work described in the paper:
  • bibliographic data: title, abstract, authors, publication venues
  • semantic descriptions of the
    • problem,
    • approach - characteristics (lossless/fuzzy, ...),
    • implementation - programming language,
    • evaluation - dataset, precision, recall, performance
  • See slide 13 in Sören's keynote at VIVO 2016 (Towards Digitizing Scholarly Communication) (here is just the slide)
• create queries or faceted browsing interface for browsing/exploring the approaches