A statistical method is suggested for data classification using the interclass correlation coefficient (ICC) statistics as a criterion function. The theory of the statistics is represented and its characteristics has ...
详细信息
A statistical method is suggested for data classification using the interclass correlation coefficient (ICC) statistics as a criterion function. The theory of the statistics is represented and its characteristics has been investigated. An algorithm for data classification using ICC is also described.< >
A rule-based expert system (ExMat) for analyzing matrices and systems of linear equation is presented. ExMat allow the user to enter a coefficient matrix and the right-hand side vector of the system. Based on rules de...
详细信息
A rule-based expert system (ExMat) for analyzing matrices and systems of linear equation is presented. ExMat allow the user to enter a coefficient matrix and the right-hand side vector of the system. Based on rules derived from the subject area of matrix theory and matrix numerical analysis, ExMat provides an expert analysis of the system. The analysis includes solvability of the systems, and the calculations of the proper inverse and solution of the system. The procedural knowledge of ExMat consists of a list of action rules determining the needed numerical procedure. The inference engine will consult the rule-based knowledge, the procedural knowledge, and then take the specified action. ExMat provides a significant improvement over standard numerical libraries in the sense that it uses its built-in analysis to select the appropriate numerical method.< >
A bound is obtained for the number of rounds of testing sufficient to identify the faulty units of a system. Within a single round each unit may participate in at most one test. The authors give an adaptive algorithm ...
详细信息
A bound is obtained for the number of rounds of testing sufficient to identify the faulty units of a system. Within a single round each unit may participate in at most one test. The authors give an adaptive algorithm which works in O(log/sub (n/t)//sup t/) rounds and uses O(n) tests. The multiplicative constants in the new bounds are small; four in both cases. This is a major improvement over previous nonadaptive and adaptive algorithm which required O(t+log n) rounds of testing and O(n+t) tests. If t>n/sup 1- epsilon /, then the algorithm runs within a constant number of rounds.< >
Many recently proposed cryptosystems are based on the assumption that factoring large composite integers is computationally difficult. In this paper we examine this assumption when the cryptanalyst has "side info...
详细信息
In this paper we present an intermediate program representation, called a program dependence graph or PDG, which summarizes not only the data dependences of each operation but also summarizes the control dependences o...
详细信息
暂无评论