The so-called(m,n)selection problem is the problem of selecting the m smallest(or largest)elements from n given numbers(n>m).With the development of parallel computers, much attention has been paid to the design of efficient algorithms of(m,n)problem for these machines.The parallel selection algorithm has been successful on networks,but seldom studied on the multiprocessing systems.This paper,based on a partitioning approach,proposes a partitioning algorithm of selection on multiprocessors using Valiant …

Rule groups with preconditions,posteonditions,and termination conditions were added to the ADVISE Meta-Expert System.Multiple,varying goals are also an attribute of the new rule groups.By treating the data collection process as separating from the rule inference engine, techniques for enhanced data acquisition were developed using semantic networks to describe relations among variables and to restructure value sets for variables dynamically.Having thus extended the ADVISE tools,an automated key to alfalfa f…

In this paper,we present a group of optimization rules to systematically improve the parallelism of non-DO loops under Leading Iteration Model.Control relations in programs are classified into three schemata,which can be conveniently represented with flow charts.Nine optimization rules are presented based on the flow chart description of Program Control Schemata,and they are used to show how to improve the parallelism of non-DO loops with examples.Finally,problems are discussed and further research efforts …

In this paper,we try to introduce a constructive definition of a computing machine,called a general Mathematical Model of Computing Machine(MMCM).As a general MMCM it should include all known MMCM as well as all unknown MMCM.We always follow two principles: (1)This general MMCM can be implemented by physical machine. (2)There is no other restriction on this general MMCM,except 1. Our conclusion is that the general MMCM is a finitely constructive and deterministic mathematical system.

In order to simulate the process of man s“thinking forward and backward”,this paper describes a new search method-forward-backward search.The relationship between this search method and two others,namely,statespace search and problem-reduction search,is also discussed.Some examples of using forward-backward search are given,from which the characteristics and efficiency of this method can be seen.

In[1],a definition of computation was given.In[3],we have clarified the necessity of the definition.Here we prove the equivalence between the definition and some models of computation.Hence the sufficiency of the definition is clear abundantly.

This paper concerns set expressions satisfying some constraints.Although algorithms processing expressions(for every kind)have been proposed in compilers,almost all of them use stacks and recursions,and adopt complicated parsing techniques.However,the simplification of set expressions can be done easily by its special characteristics.In this paper,we propose linear, nonrecursive,and thus efficient algorithms simplifying set expressions.The method can be well used in proposition calculus.

In this paper a Chinese sentence is regarded as a statement for an event and expressed with a table which has the nested structure.Seven kinds of sense coherent relationships among the events are introduced to organize relevent events into plot patterns.An understanding mechanism is given,and implementing techniques are discussed.