TY - GEN
T1 - A Purpose Driven Method for Language Comparison
AU - Brazier, F.M.
AU - van Harmelen, F.A.H.
AU - Straatman, R.
AU - Treur, J.
AU - Wijngaards, N.J.E.
AU - Willems, M.
N1 - Brazier.ea:96*6
PY - 1996
Y1 - 1996
N2 - Current efforts to compare knowledge engineering (KE) modelling languages have been limited to either rather shallow comparisons on a broad-set of languages, or to detailed comparisons with limited applicability to a narrow set of languages. In this paper we propose a novel way of organising language comparisons. This method is based on an alternating decomposition of the goals that a language tries to achieve and the linguistic methods it empIoys to achieve these goals. This new method for comparing languages allows a general comparison at high levels of abstraction, while not preventing more precise comparisons whenever possible. One result of our comparison method is an insight in the different assumptions that underly the languages to be compared. Two further consequences follow from the proposed comparison method, namely (i) a measure for the degree of similarity between languages, and (ii) a method for translating between languages. After describing our method, we apply it to a pair of KE modelling languages, and show how it yields insights in the assumptions underlying the languages and how it can be used to produce a translation procedure between the languages.
AB - Current efforts to compare knowledge engineering (KE) modelling languages have been limited to either rather shallow comparisons on a broad-set of languages, or to detailed comparisons with limited applicability to a narrow set of languages. In this paper we propose a novel way of organising language comparisons. This method is based on an alternating decomposition of the goals that a language tries to achieve and the linguistic methods it empIoys to achieve these goals. This new method for comparing languages allows a general comparison at high levels of abstraction, while not preventing more precise comparisons whenever possible. One result of our comparison method is an insight in the different assumptions that underly the languages to be compared. Two further consequences follow from the proposed comparison method, namely (i) a measure for the degree of similarity between languages, and (ii) a method for translating between languages. After describing our method, we apply it to a pair of KE modelling languages, and show how it yields insights in the assumptions underlying the languages and how it can be used to produce a translation procedure between the languages.
UR - http://www.scopus.com/inward/record.url?scp=84958767794&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/record.url?scp=84958767794&partnerID=8YFLogxK
U2 - 10.1007/3-540-61273-4_5
DO - 10.1007/3-540-61273-4_5
M3 - Conference contribution
SN - 3540612734
SN - 9783540612735
VL - 1076
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 66
EP - 81
BT - Advances in Knowledge Acquisition - 9th European Knowledge Acquisition Workshop, EKAW 1996, Proceedings
A2 - Shadbolt, N.
A2 - O'Hara, K.
A2 - Schreiber, A.Th.
PB - Springer/Verlag
T2 - 9th European Knowledge Acquisition Workshop, EKAW 1996
Y2 - 14 May 1996 through 17 May 1996
ER -