为有效提高C语言中间表达式解读程序信息的能力,在结合C语言自身语言特点的基础上,引入流程控制图(Flow Control Graph,FCG)匹配自动评分方法,该方法能够根据C语言的特点,计算学生答案与参考答案之间的相似度,给出评分。引入最近邻(Flow...为有效提高C语言中间表达式解读程序信息的能力,在结合C语言自身语言特点的基础上,引入流程控制图(Flow Control Graph,FCG)匹配自动评分方法,该方法能够根据C语言的特点,计算学生答案与参考答案之间的相似度,给出评分。引入最近邻(Flow Control-KNN,FC-KNN)算法来对FCG算法进行模板脱敏,在FCG的基础上,运用k临近算法根据提取的特征对程序进行评分。实验结果证明,FCG和FC-KNN算法在独立运行时分别具有91.5%和92.3%的平均准确率,而经过融合后,算法之间实现了优势互补,准确率提升到94.0%,在独立运行的情况下,FC-KNN算法的评分效果较FCG好,准确性更高,对两种算法进行数据融合、优势互补,验证了集成后的分类模型在评分的整个过程中,均能够达到良好的分类效果,具有较高的准确率。展开更多
To solve the problems that the exception handling code is hard to test and maintain and that it affects the robustness and reliability of software, a method for evaluating the exception handling of programs is present...To solve the problems that the exception handling code is hard to test and maintain and that it affects the robustness and reliability of software, a method for evaluating the exception handling of programs is presented. The exception propagation graph (EPG) that describes the large programs with exception handling constructs is proposed by simplifying the control flow graph and it is applied to a case to verify its validity. According to the EPG, the exception handling code that never executes is identified; the points that are the most critical to controlling exception propagation are found; and the irrational exception handling code is corrected. The constructing algorithm for the EPG is given; thus, this provides a basis for automatically constructing the EPG and automatically correcting the irrational exception handling code.展开更多
On the basis of software testing tools we developed for programming languages, we firstly present a new control flowgraph model based on block. In view of the notion of block, we extend the traditional program\|based ...On the basis of software testing tools we developed for programming languages, we firstly present a new control flowgraph model based on block. In view of the notion of block, we extend the traditional program\|based software test data adequacy measurement criteria, and empirically analyze the subsume relation between these measurement criteria. Then, we define four test complexity metrics based on block. They are J\|complexity 0; J\|complexity 1; J\|complexity \{1+\}; J\|complexity 2. Finally, we show the Kiviat diagram that makes software quality visible.展开更多
Today's automation industry is driven by the need for an increased productivity, higher flexibility, and higher individuality, and characterized by tailor-made and more complex control solutions. In the processing in...Today's automation industry is driven by the need for an increased productivity, higher flexibility, and higher individuality, and characterized by tailor-made and more complex control solutions. In the processing industry, logic controller design is often a manual, experience-based, and thus an error-prone procedure. Typically, the specifications are given by a set of informal requirements and a technical flowchart and both are used to be directly translated into the control code. This paper proposes a method in which the control program is constructed as a sequential function chart (SFC) by transforming the requirements via clearly defined intermediate formats. For the purpose of analysis, the resulting SFC can be translated algorithmically into timed automata. A rigorous verification can be used to determine whether all specifications are satisfied if a formal model of the plant is available which is then composed with the automata model of the logic controller (LC).展开更多
文摘为有效提高C语言中间表达式解读程序信息的能力,在结合C语言自身语言特点的基础上,引入流程控制图(Flow Control Graph,FCG)匹配自动评分方法,该方法能够根据C语言的特点,计算学生答案与参考答案之间的相似度,给出评分。引入最近邻(Flow Control-KNN,FC-KNN)算法来对FCG算法进行模板脱敏,在FCG的基础上,运用k临近算法根据提取的特征对程序进行评分。实验结果证明,FCG和FC-KNN算法在独立运行时分别具有91.5%和92.3%的平均准确率,而经过融合后,算法之间实现了优势互补,准确率提升到94.0%,在独立运行的情况下,FC-KNN算法的评分效果较FCG好,准确性更高,对两种算法进行数据融合、优势互补,验证了集成后的分类模型在评分的整个过程中,均能够达到良好的分类效果,具有较高的准确率。
基金The National Natural Science Foundation of China(No60503020)the National Basic Research Program of China (973Program) (No2002CB312000)+1 种基金the Natural Science Foundation of Jiangsu Province (NoBK2006094)the Science Research Foundation of China University of Mining and Technology
文摘To solve the problems that the exception handling code is hard to test and maintain and that it affects the robustness and reliability of software, a method for evaluating the exception handling of programs is presented. The exception propagation graph (EPG) that describes the large programs with exception handling constructs is proposed by simplifying the control flow graph and it is applied to a case to verify its validity. According to the EPG, the exception handling code that never executes is identified; the points that are the most critical to controlling exception propagation are found; and the irrational exception handling code is corrected. The constructing algorithm for the EPG is given; thus, this provides a basis for automatically constructing the EPG and automatically correcting the irrational exception handling code.
文摘On the basis of software testing tools we developed for programming languages, we firstly present a new control flowgraph model based on block. In view of the notion of block, we extend the traditional program\|based software test data adequacy measurement criteria, and empirically analyze the subsume relation between these measurement criteria. Then, we define four test complexity metrics based on block. They are J\|complexity 0; J\|complexity 1; J\|complexity \{1+\}; J\|complexity 2. Finally, we show the Kiviat diagram that makes software quality visible.
基金the European Union through the Network of Excellence Hybrid Control (HYCON) under contract IST-511368.
文摘Today's automation industry is driven by the need for an increased productivity, higher flexibility, and higher individuality, and characterized by tailor-made and more complex control solutions. In the processing industry, logic controller design is often a manual, experience-based, and thus an error-prone procedure. Typically, the specifications are given by a set of informal requirements and a technical flowchart and both are used to be directly translated into the control code. This paper proposes a method in which the control program is constructed as a sequential function chart (SFC) by transforming the requirements via clearly defined intermediate formats. For the purpose of analysis, the resulting SFC can be translated algorithmically into timed automata. A rigorous verification can be used to determine whether all specifications are satisfied if a formal model of the plant is available which is then composed with the automata model of the logic controller (LC).