| | |
Summary: Fig. 1. The conversational case retrieval process in CCBR
CaseQuery := Case-Query-formalize(InitialProblemDescription);
Repeat:
CaseQuery:=Dialogue-Inference(CaseQuery);
RankedRetrievedCases := CBR-Retrieve(CaseQuery);
UnknownQuestions := Question-Identify(RankedRetrievedCases, CaseQuery);
RankedUnknownQuestions := Question-Rank(UnknownQuestions);
Display(RankedUnknownQuestions, RankedRetrievedCases);
If (users find their desired cases or have no question to answer) then
Exit loop;
Else
UserAnswer := User-Select-and-Answer-Question();
CaseQuery :=Case-Query-Update(CaseQuery, UserAnswer);
Explanation-Boosted Question Selection in
Conversational CBR
Mingyang Gu and Agnar Aamodt
Department of Computer and Information Science, Norwegian University of Science and
Technology, Sem Sælands vei 7-9, N-7491, Trondheim, Norway
{Mingyang.Gu, Agnar.Aamodt}@idi.ntnu.no
Abstract. A core research concern in conversational case-based reasoning
|
