Title :
Improved filtering for the QSIM algorithm
Author :
Say, A. C Cem ; Kuru, Selahattin
Author_Institution :
Dept. of Comput. Eng., Bogazici Univ., Istanbul, Turkey
fDate :
9/1/1993 12:00:00 AM
Abstract :
A source of spurious predictions inside the qualitative simulation algorithm QSIM proposed by B. Kuiper (1986) is identified. The proposed solution involves the use of interval corresponding values for filtering inconsistent states and does not require any additions or restrictions in the input set, unlike the other approaches to the spurious prediction elimination problem. The time and space complexities of the algorithm are not affected by the modification
Keywords :
filtering and prediction theory; inference mechanisms; simulation; QSIM; filtering; inconsistent states; interval corresponding values; qualitative simulation algorithm; space complexities; spurious predictions; time complexities; Arithmetic; Artificial intelligence; Filtering algorithms; Law; Legal factors; Machine intelligence; Predictive models;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on