Solution reuse for textual cases

The reuse stage in textual CBR identifies reusable textual constructs in solution content. This involves content annotation so that reusable solution text is easily identifiable from the rest. We present a generic architecture, Case Retrieval Reuse Net (CR2N), that can be used to annotate text to denote text content as reusable or not. Initial results from a weather forecast revision dataset shows up to 80% accuracy and a significantly better precision over a retrieve-only system. Although our work with CR2N is still on going, it also provides useful insight into the text revision stage. Keyword Textual CBR, Case Reuse, Case Retrieval Net, NLG