Standardising the elimination of unnecessary data from interview transcripts for the purposes of knowledge‐based systems — a case study

Abstract

Although knowledge elicitation, the process of extracting knotvledgefrom human experts to be incorporated into a knowledge-basedsystem, has been the subject of some notable studies, less attention has been paid to the methods of analysing the raw data once it has been extractedfrom the expert. When knowledge elicitation sessions are interview-based, the resultant form of raw data is usually a transcript of the interviewee's utterances. This paper describes an investigation into the preliminary stage of analysing such transcripts. It outlines the development of an approach to eliminate unnecessary detailfrom interview transcripts, thus enabling attention to be focused upon the remaining, more relevant data via a simple technique based upon cheap and readily available technology. The paper then outlines a rapid-prototyping approach for evaluating this method, the results of which werefelt to be very encouraging.

A comparison of knowledge elicitation methods

“…The main emphasis of this section will therefore be to focus upon the methods by which this already elicited data was compared, rather than to provide an account of the actual KE sessions themselves, these accounts already having been made available (in Jones et al 1996a andJones et al 1996b). …”

“…As previously mentioned, the main focus of this paper rests upon the comparison of two very different techniques which have been used for KE purposes, using knowledge already extracted from a human expert. The main emphasis of this section will therefore be to focus upon the methods by which this already elicited data was compared, rather than to provide an account of the actual KE sessions themselves, these accounts already having been made available (in Jones et al 1996a andJones et al 1996b).…”

“…When an attempt was made to analyse the S-DEM and NS-DEM derived transcripts using the techniques described in Jones et al (1996a), the results were not as satisfactory, and were considerably more difficult to interpret. This was largely due to the different circumstances under which the analysis techniques had been previously used, i.e.…”

“…The first attempt at analysis was made via methods which the authors had used previously with considerable success (see Jones et al 1996a), and which were, in fact, instrumental in the construction of the S-DEM and NS-DEM systems themselves. Full details of the analytic techniques are not included here, due to their being considered peripheral to the main issue of comparing methods of elicitation.…”

“…and Jones et al (1996a), that rhetorical questioning can be considered to be an external indication that significant cognitive activity is taking place. This issue is not dealt with at length here due to it being peripheral to the main theme, a fuller account of this phenomena is available in Jones et al(1 996a).…”