In recent years, products have become more complex due to increased and more sophisticated requirements, and development time has become shorter to respond to changes in the market. Therefore, a reusable system is important for the efficient use of simulation. In this paper, we propose patterns for reusable simulation languages and how to apply them. This pattern divides the simulation target into State and Actions, and relates with a sum of zero between State and Actions. Furthermore, Strategy pattern is applied to increase reusability during Action development. We also apply this pattern using Modelica language, an object-oriented physical modeling language. By applying these patterns, the new reusable simulation library can now be created from scratch.
When evaluating a beginners’ ability in programming, it is a standard method to have them write programs to solve problems and judge their correctness. To apply this method in examinations that are taken by a large number of people, it is necessary to automatically evaluate the correctness of the program. In the method of preparing inputs and their expected outputs and comparing them with the outcome of the submitted program, an incorrect program may be mistakenly judged as correct. In this study, we propose a programming language based on Presburger arithmetic that can automatically judge the correctness of any program written in the language. In addition, to alleviate the difficulty that beginners tend to stumble due to the necessary restrictions for realizing automatic scoring, we introduce an environment in which only programs that conform to the restrictions can be created by combining blocks. It was confirmed that the answers to some typical questions can be created and automatically scored.
In smartphone and Web applications, a major problem arises when the rendering result of a GUI component does not meet a defined specification. In this study, we refer to such problems as Presentation Failure, and although many methods have been proposed to detect Presentation Failure, most of them are achieved by directly comparing images of the application output with the specified specifications. Therefore, a large difference between the two may induce misdetections of Presentation Failure. This study proposes a method to extract abstract visual properties from the output of before and after version web pages and compare visual properties extracted correspondingly. Specifically, screenshots are taken from each image for each element, and visual properties are extracted using computer vision techniques. Furthermore, we implement the proposed method as a tool (SpiderTailed) and describe the comparison experiment results with a manual observation test to confirm the effectiveness of the method.
The Sustainable Development Goals (SDGs) initiative by corporations requires that their business activities be linked to the 17 SDGs. Moreover, the digital transformation (DX) of enterprises requires the use of digital technology to realize a digital company with competitive advantage. It is inefficient to implement the SDGs and DX initiatives separately in an enterprise. In this paper, we propose a solution to the question, “How do we combine knowledge for the SDGs and DX?”
A system specification is written in a natural language, while a formal specification is useful for discovering faults by using formal techniques, such as model checking. Consequently, analysis techniques based on natural language processing are desirable for translating system specifications into formal specifications. The matching system specifications with formal specifications is the necessary stage when extracting formal specifications. However, a generic and accurate matching method has not yet been established. In this paper, we propose a pattern matching method for identifying the specification of sentences. Specifically, we constructed a semantic analyzer that transforms sentences into logical expressions that entail patterns based on event semantics. We also applied pattern matching to practical requirement statements. Based on the simple pattern matching, all sentences entailed several patterns. The results of the entailment recognition showed that it was possible to distinguish between sentences that represent the state transition and those that do not. The results are valid for formal specifications.
