An Evolutionary Algorithm to Improve Knowledge-Based Decision-Making

Expert system decision making has proved effective in guiding the process of automatic transformation of sequential legacy codes to parallel equivalents for execution on multiprocessor computer systems. Parallelization decisions normally taken by human experts are replaced by those influenced by domain-specific and ‘history’ casebased knowledge. However, the information obtained from the knowledge base can be both limited and unreliable, reducing the effectiveness of the decision making process. A time-consuming iterative trial-and-error process is required in order to find the transformation to obtain the best possible performance results. This paper introduces a technique which employs a genetic algorithm to guide the process of performance-related solution-finding, while reducing the penalties inherent with exhaustive testing. Key-Words: Automatic, Intelligent, Parallelisation, Distribution, Expert-Systems, Knowledge-Based, Multiprocessor, Evolutionary-Algorithm, Genetic-Algorithm, Case-Based-Reasoning

[1]  Pablo Moscato,et al.  On Evolution, Search, Optimization, Genetic Algorithms and Martial Arts : Towards Memetic Algorithms , 1989 .

[2]  Peter Milligan,et al.  A knowledge based approach to parallel software engineering , 1996, Software Engineering for Parallel and Distributed Systems.

[3]  S. P. Johnson,et al.  Integrating User Knowledge with Information from Parallelisation Tools to Facilitate the Automatic Generation of Efficient Parallel FORTRAN Code , 1996, Parallel Comput..

[4]  Barbara M. Chapman,et al.  Automatic Support for Data Distribution on Distributed Memory Multiprocessor Systems , 1993, LCPC.

[5]  Ibrahim Zeid,et al.  DEJAVU: Case-based reasoning for mechanical design , 1993, Artif. Intell. Eng. Des. Anal. Manuf..

[6]  C. S. Krishnamoorthy,et al.  A framework for case-based reasoning in engineering design , 1995, Artificial Intelligence for Engineering Design, Analysis and Manufacturing.

[7]  Ray W. Clough,et al.  Convergence Criteria for Iterative Processes , 1972 .

[9]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[10]  Edmund K. Burke,et al.  Initialization Strategies and Diversity in Evolutionary Timetabling , 1998, Evolutionary Computation.

[11]  D. E. Goldberg,et al.  Genetic Algorithms in Search , 1989 .

[12]  HamadaKazuro,et al.  Hybridizing a Genetic Algorithm with Rule-Based Reasoning for Production Planning , 1995 .

[13]  Barry McCollum,et al.  Knowledge Exploitation for Improved Data Distribution , 1999, PaCT.

[14]  Ken-ichi Sato,et al.  Hybridizing a Genetic Algorithm with Rule-Based Reasoning for Production Planning , 1995, IEEE Expert.