扩展模糊描述逻辑是对描述逻辑的一种模糊扩展,支持对复杂模糊知识的表示和推理,但该逻辑缺乏支持术语公理约束的推理算法.提出扩展模糊描述逻辑EFALCR+(extended fuzzy attributive concept description language with complements and...扩展模糊描述逻辑是对描述逻辑的一种模糊扩展,支持对复杂模糊知识的表示和推理,但该逻辑缺乏支持术语公理约束的推理算法.提出扩展模糊描述逻辑EFALCR+(extended fuzzy attributive concept description language with complements and transitive roles)的受限TBox(terminological box)描述术语公理,给出受限TBox约束下的EFALCR+推理算法,并对该算法进行优化,证明优化后的算法是正确完备的,时间复杂性不超过指数,最后证明受限TBox约束下的EFALCR+推理问题是指数时间完全问题.优化算法的最坏时间复杂性已达到该问题推理算法的复杂度下界,是实现术语公理约束下模糊知识库推理的有效算法.展开更多
语义Web模糊知识的表示和应用常常涉及模糊隶属度比较,但现有描述逻辑的模糊扩展缺乏描述模糊隶属度比较的能力.提出支持模糊隶属度比较和描述逻辑ALCN(attributive concept description language with complements and number restrict...语义Web模糊知识的表示和应用常常涉及模糊隶属度比较,但现有描述逻辑的模糊扩展缺乏描述模糊隶属度比较的能力.提出支持模糊隶属度比较和描述逻辑ALCN(attributive concept description language with complements and number restriction)概念构造子的扩展模糊描述逻辑FCALCN(fuzzy comparable ALCN).FCALCN引入新的原子概念形式以支持模糊隶属度比较.给出FCALCN的推理算法,证明了在空TBox约束下FCALCN的推理问题复杂性是多项式空间完全的.FCALCN能够表达语义Web上涉及模糊隶属度比较的复杂模糊知识并实现对它们的推理.展开更多
Some metamorphic relations （MR） are not good at detecting faults in metamorphic testing. In this paper, the method of making compositional MR （CMR） based on the speculative law of proposition logic is presented. T...Some metamorphic relations （MR） are not good at detecting faults in metamorphic testing. In this paper, the method of making compositional MR （CMR） based on the speculative law of proposition logic is presented. This method constructs new MRs by composing existing MRs in a pairwise way. Because CMR contains all the advantages of the MRs that form it, its fault detection performance is wonderful. On the other hand, the number of relations will decrease greatly after composing, so a program can be tested with much fewer test cases when CMRs are used. In order to research the characteristics of a CMR, two case studies are analyzed. The experimental results show that the CMR＇s performance is mostly determined by the central MRs forming it and the sequence of composition. Testing efficiency is improved greatly when CMRs are used.展开更多
文摘扩展模糊描述逻辑是对描述逻辑的一种模糊扩展,支持对复杂模糊知识的表示和推理,但该逻辑缺乏支持术语公理约束的推理算法.提出扩展模糊描述逻辑EFALCR+(extended fuzzy attributive concept description language with complements and transitive roles)的受限TBox(terminological box)描述术语公理,给出受限TBox约束下的EFALCR+推理算法,并对该算法进行优化,证明优化后的算法是正确完备的,时间复杂性不超过指数,最后证明受限TBox约束下的EFALCR+推理问题是指数时间完全问题.优化算法的最坏时间复杂性已达到该问题推理算法的复杂度下界,是实现术语公理约束下模糊知识库推理的有效算法.
文摘语义Web模糊知识的表示和应用常常涉及模糊隶属度比较,但现有描述逻辑的模糊扩展缺乏描述模糊隶属度比较的能力.提出支持模糊隶属度比较和描述逻辑ALCN(attributive concept description language with complements and number restriction)概念构造子的扩展模糊描述逻辑FCALCN(fuzzy comparable ALCN).FCALCN引入新的原子概念形式以支持模糊隶属度比较.给出FCALCN的推理算法,证明了在空TBox约束下FCALCN的推理问题复杂性是多项式空间完全的.FCALCN能够表达语义Web上涉及模糊隶属度比较的复杂模糊知识并实现对它们的推理.
基金The National Natural Science Foundation of China(No.60425206,60633010,60773104,60503033)the Excellent Talent Foundation of Teaching and Research of Southeast University
文摘Some metamorphic relations （MR） are not good at detecting faults in metamorphic testing. In this paper, the method of making compositional MR （CMR） based on the speculative law of proposition logic is presented. This method constructs new MRs by composing existing MRs in a pairwise way. Because CMR contains all the advantages of the MRs that form it, its fault detection performance is wonderful. On the other hand, the number of relations will decrease greatly after composing, so a program can be tested with much fewer test cases when CMRs are used. In order to research the characteristics of a CMR, two case studies are analyzed. The experimental results show that the CMR＇s performance is mostly determined by the central MRs forming it and the sequence of composition. Testing efficiency is improved greatly when CMRs are used.
