Yoshimichi Watanabe

A method of program refactoring based on code clone detection and impact analysis

This paper proposes a method that aggregates similar portions in a program into one common portion and redesigns current program structure to appropriate program structure. When a new function is added to an existing program, the function tends to be developed by copying & pasting a portion in the program and modifying its portion. As a result, it becomes to exist many similar portions in the program. In the case that error modifications or changes occurs in the similar portion, appropriate modifications are required to the all similar portions. It would be considered that the quality and efficiency of those tasks are decreased. So that, this paper proposes a method that detect similar portions with minor modifications and a method that aggregates those similar portions to one appropriate common portion with well-defined program structure. As a result, current program becomes to be refined to the program that can accommodate future modification or changes properly.

A Method for Software Requirement Volatility Analysis Using QFD

Changes of software requirements are inevitable during the development life cycle. Rather than avoiding the circumstance, it is easier to just accept it and find a way to anticipate those changes. This paper proposes a method to analyze the volatility of requirement by using the Quality Function Deployment (QFD) method and the introduced degree of volatility. Customer requirements are deployed to software functions and subsequently to architectural design elements. And then, after determining the potential for changes of the design elements, the degree of volatility of the software requirements is calculated. In this paper the method is described using a flow diagram and illustrated using a simple example, and is evaluated using a case study.

Maintaining data consistency of XML databases using verification techniques

XML document is a very popular way to transfer data between databases because of its simplicity of structure and text based data storage. In some typical implementation scenarios where XML document is used to transfer data as XML database, it is necessary to check the consistency of the data at destination. If users are responsible for storing and transferring the document, there is a possibility of data manipulation. In this paper, we propose to apply data verification techniques including checksum and hash algorithms for data verification of XML document. As an advantage, we show that we not only can check the documents authenticity, but also we can identify the specific modified portions of the document. We discuss many popular checksum techniques and secured hash algorithms applied to XML document and compare their efficiency on data verification in the scenarios of the XML data transfer.