| 摘要 |
<p>A proposition X1 = PL which appears on the highest frequency in the condition parts of a group of fuzzy inference rules is extracted and one group is formed with rules (1, 2, 6, 8, and 9) including this proportion. The rule (8) including only the proposition X1 = PL in its condition part is placed at the head of the group. A proposition X2 = PM which appears on the second highest frequency in the residual rules (1, 2, 6 and 9) is extracted and a subgroup is formed with the rules (1, 6, and 9) including this proposition. The rules of the subgroup arranged following the rule (8) in a specific order. The residual rule (2) is arranged following the rules of the subgroup. Other rules (3, 4, 5, 7, and 10) are processed similarly. In this manner, the overlapping processings relative to overlapping propositions are avoided and the improvement of efficiency is attained by rearranging the rules according to the frequencies on which the propositions included in the rules appear. <IMAGE></p> |
