研究者詳細 - 片山　徹郎

Improvement of RETUSS to Ensure Traceability between Sequence Diagram in UML and Java Source Code in Real Time 査読あり

Arima Kaoru, Katayama Tetsuro, Kita Yoshihiro, Yamaba Hisaaki, Aburada Kentaro, Okazaki Naonobu

ICAROB 2021: PROCEEDINGS OF THE 2021 INTERNATIONAL CONFERENCE ON ARTIFICIAL LIFE AND ROBOTICS 352 - 355 2021年1月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

The Seven Information Features of Class for Blob and Feature Envy Smell Detection in a Class Diagram 査読あり

Bayu Priyambadha, Katayama Tetsuro, Kita Yoshihiro, Yamaba Hisaaki, Aburada Kentaro, Okazaki Naonobu

ICAROB 2021: PROCEEDINGS OF THE 2021 INTERNATIONAL CONFERENCE ON ARTIFICIAL LIFE AND ROBOTICS 348 - 351 2021年1月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Test Suite Reusability Measurement Based on Uses Frequently and Coverage of Reused Test Cases 査読あり

Mochamad Chandra Saputra, Katayama Tetsuro, Kita Yoshihiro, Yamaba Hisaaki, Aburada Kentaro, Okazaki Naonobu

ICAROB 2021: PROCEEDINGS OF THE 2021 INTERNATIONAL CONFERENCE ON ARTIFICIAL LIFE AND ROBOTICS 344 - 347 2021年1月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Identification of Redundant Test Cases by Using Similarity Measurement Method for Test Suite Evaluation 査読あり

Saputra Mochamad Chandra, Katayama Tetsuro, Kita Yoshihiro, Yamaba Hisaaki, Aburada Kentaro, Okazaki Naonobu

Journal of Advances in Artificial Life Robotics 1 ( 4 ) 187 - 192 2021年

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：株式会社 ALife Robotics

Evaluating the test suite that contains redundant test cases is necessary to ensure the efficiency of the testing and reducing the cost of testing. The principle of redundant test cases on this research is any test cases on a similar path executed with a similar high value of normalized Euclidean distance. The first, the similarity calculated between the test cases and path coverage uses Euclidean distance. The value of the Euclidean distance on the test case with the lowest value of distance indicating highly similar and possibly executing similar lines of code. The normalized Euclidean distance is using to normalize the value from Euclidean distance result. The experiment uses two java programs. Each redundancy score is 0.37 and 0.67, respectively. It means 37% and 67% of the test cases should be avoided because there are inefficiency test cases on the test suite. The research confirms redundant test cases can be identified by Euclidean distance and normalized Euclidean distance to evaluate the test suite.

DOI： 10.57417/jaalr.1.4_187

Extension of the Function to Ensure Real-time Traceability between UML Sequence Diagram and Java Source Code on RETUSS 査読あり

Arima Kaoru, Katayama Tetsuro, Kita Yoshihiro, Yamaba Hisaaki, Aburada Kentaro, Okazaki Naonobu

Journal of Advances in Artificial Life Robotics 2 ( 1 ) 37 - 41 2021年

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：株式会社 ALife Robotics

Ensuring traceability of software deliverables is one of the methods to ensure software quality. RETUSS (Real-time Ensure Traceability between UML and Source-code System) is a tool that saves labor and time, and eliminates mistakes by human handling in ensuring traceability between UML and source code. However, RETUSS is not useful due to its limited scope of application. This paper improves the usefulness of RETUSS by extending the function to ensure real-time traceability between UML sequence diagrams and Java source code on RETUSS.

DOI： 10.57417/jaalr.2.1_37

Proposal of an algorithm to generate VDM++ specification based on its grammar by using word lists extracted from the natural language specification 査読あり

Katayama T., Shigyo Y., Kita Y., Yamaba H., Aburada K., Okazaki N.

Journal of Robotics, Networking and Artificial Life 7 ( 3 ) 165 - 169 2020年12月

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：Journal of Robotics, Networking and Artificial Life

DOI： 10.2991/jrnal.k.200909.005

Code Coverage Similarity Measurement Using Machine Learning for Test Cases Minimization 査読あり

Saputra M.C., Katayama T.

2020 IEEE 9th Global Conference on Consumer Electronics, GCCE 2020 287 - 291 2020年10月

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：2020 IEEE 9th Global Conference on Consumer Electronics, GCCE 2020

© 2020 IEEE. Machine learning approach for minimizing the number of test cases on the test suite is an interesting research area on software testing. The research tries to minimize the number of test cases on the test suite by minimizing redundant test cases based on similarity classification. The Support Vector Machine, K-Nearest Neighbour, and Decision tree classify similar test cases by comparing the lines executed by test cases. The result has shown that the support vector machine is the highest score on accuracy and the lowest score on error rate comparing with K-Nearest Neighbour, and Decision tree. Minimize the redundant test cases increase the quality of the test cases, and reducing time on the testing process.

DOI： 10.1109/GCCE50665.2020.9291990

Tree-based keyword search algorithm over the visual paradigm's class diagram xml to abstracting class information 査読あり

Priyambadha B., Katayama T.

2020 IEEE 9th Global Conference on Consumer Electronics, GCCE 2020 280 - 284 2020年10月

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：2020 IEEE 9th Global Conference on Consumer Electronics, GCCE 2020

© 2020 IEEE. In software engineering, abstracting the class information from XML based design artifact is important to analyze the design phase. The challenge of this research is the information is embedded inner the XML data structure and the target data is more than one and stored in a non-consecutive XML element position. This paper proposes a search algorithm over the XML file using the tree-based keyword as a composition of information. The experimentation confirms that the proposed algorithm can be used to collect all information that lies on the XML file by using the tree-based keyword consistently. The proposed algorithm's performance shows the linear growth rate of the execution time that inline with the growth of the number of lines. The tree-based keyword searching algorithm is promising used in the analysis of XML based software artifact.

DOI： 10.1109/GCCE50665.2020.9291865

Proposal of an Approach to Generate VDM++ Specifications from Natural Language Specification by Machine Learning 査読あり

Shigyo Y., Katayama T.

2020 IEEE 9th Global Conference on Consumer Electronics, GCCE 2020 292 - 296 2020年10月

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：2020 IEEE 9th Global Conference on Consumer Electronics, GCCE 2020

© 2020 IEEE. A natural language contains ambiguous expressions. The VDM++ is one of the methodotogies on the formal methods to write the specification without ambiguity. It is difficult to write a VDM++ specification, because VDM++ is written by strict grammar. This research proposes an approach to automatically generate the VDM++ specification by machine learning. This approach defines four data structures and has four processes. In this paper, variables and only real type in the VDM++ specification are generated automatically by this approach. In order to generate the variables and real type, it is necessary to extract the noun corresponding to the variable from the natural language specification. Consequently, our proposed approach can generate a VDM++ specification and we have confirmed that the generated VDM++ specification is grammatically correct.

DOI： 10.1109/GCCE50665.2020.9292047

機械学習を用いて自然言語仕様書から生成した分類リストを用いたVDM++仕様書生成アプローチの提案査読あり

執行泰弘, 片山徹郎

宮崎大学工学部紀要 49 245 - 250 2020年9月

記述言語：日本語掲載種別：研究論文（学術雑誌）出版者・発行元：宮崎大学工学部

Specifications are generally written in natural language. Natural language contains ambiguity. As a method of writing a specification without ambiguity, VDM which is a formal method exists. Because it is difficult to write specification languages such as VDM++ because they have strict grammars data types and system invariants that are not found in natural language specifications. This study attempts to generate automatically a VDM++ specification from the natural language specification by using machine learning. For automatic generation of VDM++, it is necessary to extract predicates corresponding to the function names and nouns corresponding to variable names from the natural language specification. However,it is difficult to generate a VDM++ specification by using only the extracted nouns and predicates. This paper proposes an approach to generate automatically a VDM++ specification from extracted words list. An identifier is generated from the extracted words, and the VDM++ specification can be generated by converting this identifier into a VDM++ grammar.

主観的輪郭を用いた回転3Dテキスト型CAPTCHAの提案査読あり

臼崎翔太郎, 油田健太郎, 山場久昭, 片山徹郎, 椋木雅之, 朴美娘i, 岡崎直宣

電子情報通信学会論文誌 B J103-B ( 8 ) 332 - 343 2020年8月

記述言語：日本語掲載種別：研究論文（学術雑誌）出版者・発行元：電子情報通信学会

テキスト型CAPTCHAは機械の攻撃精度が高く，実用には複雑な妨害が必要である．本研究では，主観的輪郭に着目し，複雑な妨害が必要ないテキスト型CAPTCHAを検討する．本手法では3D空間に縦か横に回転する3D文字を配置し，指定された向きに回転する文字を入力させる．3D文字は側面部のみが視認されるようにしており，人間は主観的輪郭の知覚により側面部分の形状から文字を判読できるが，機械には，配置位置や向きで異なる形状を同じ文字だと認識するのは困難である．また回転運動は，人間には文字認識の手がかりとなるが，機械には回転方向の機械的な認識も必要となる．評価実験の結果，人間成功率は0.91であり，既存のテキスト型CAPTCHAの平均成功率を4ポイント上回った．深層学習手法のCNNによる攻撃プログラムを作成し計1,000問のCAPTCHAに適用した結果，文字の認識率は0.930，機械成功率は0.618であり，今後は他の妨害を検討する必要があることが分かった．またユーザビリティ評価では回答時間とSUSを調査した．平均回答時間は21.4秒と既存手法よりも8.4秒長いが，SUSは73.3であり，画像型CAPTCHAと同等程度の高いユーザビリティを保持していることが分かった．

DOI： 10.14923/transcomj.2019WFP0007

Considering the Semantic Approach to Assess the Class Cohesion 招待あり

Bayu Priyambadha, T. Katayama

Japan Symposium on Software Testing 2020 Tokyo (JaSST'20 Tokyo) 236 - 246 2020年3月

担当区分：責任著者記述言語：英語掲載種別：研究論文（研究会，シンポジウム資料等）

Test Case Redundancy Elimination Uses Similarity Measurement on Quality Assurance Testing 招待あり

Mochamad Chandra Saputra, T. Katayama

Japan Symposium on Software Testing 2020 Tokyo (JaSST'20 Tokyo) 224 - 235 2020年3月

担当区分：責任著者記述言語：英語掲載種別：研究論文（研究会，シンポジウム資料等）

Evaluation of Yubimoji Based Gestures for Realizing User Authentication Method Using s-EMG 査読あり

Yamaba H., Nagatomo Y., Usuzaki S., Takatsuka K., Aburada K., Katayama T., Park M., Okazaki N.

Advances in Intelligent Systems and Computing 1150 AISC 834 - 844 2020年3月

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：Advances in Intelligent Systems and Computing

© 2020, Springer Nature Switzerland AG. At the present time, since mobile devices such as tablet-type PCs and smart phones have penetrated deeply into our daily lives, an authentication method that prevents shoulder surfing attacks comes to be important. We are investigating a new user authentication method for mobile devices that uses surface electromyogram (s-EMG) signals, not screen touching. The s-EMG signals, which are generated by the electrical activity of muscle fibers during contraction, can be detected over the skin surface, and muscle movement can be differentiated by analyzing the s-EMG signals. Taking advantage of the characteristics, we proposed a method that uses a list of gestures as a password in the previous study. In order to realize this method, we have to prepare a sufficient number of gestures that are used to compose passwords. In this paper, we adopted fingerspelling as candidates of such gestures. We introduced manual kana of the Japanese Sign Language syllabary and selected the candidate gestures based on them. Their performances were evaluated by constructing their identifier using support vector machines.

DOI： 10.1007/978-3-030-44038-1_76

Test Cases Redundant Elimination on Code Coverage Uses Distance and Correlation Measurement Method 査読あり

Mochamad Chandra Saputra, Tetsuro Katayama, Yoshihiro Kita, Hisaaki Yamaba, Kentaro Aburada, Nanoubu Okazaki

PROCEEDINGS OF THE 2020 INTERNATIONAL CONFERENCE ON ARTIFICIAL LIFE AND ROBOTICS (ICAROB2020) 755 - 758 2020年1月

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：ALIFE ROBOTICS CO, LTD

The Euclidean distance and correlation measured by comparing the line executed by the test case to find the similarity of the test cases. The test cases have the lowest value of distance means highly similar and possible executing a similar line or path. The research tries to eliminate redundant test cases based on that similarity. Several redundant test cases are eliminated to get the best test cases. By Euclidean distance, the research can eliminate a similar test case on the test suite.

The Measurement of Class Cohesion using Semantic Approach 査読あり

Bayu Priyambadha, Tetsuro Katayama, Yoshihiro Kita, Hisaaki Yamaba, Kentaro Aburada, Nanoubu Okazaki

PROCEEDINGS OF THE 2020 INTERNATIONAL CONFERENCE ON ARTIFICIAL LIFE AND ROBOTICS (ICAROB2020) 759 - 762 2020年1月

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：ALIFE ROBOTICS CO, LTD

The cohesion is one of the design quality indicators that can be measured by focusing on the correlation between elements inner the class. In Direct Distance Design Class Cohesion (D3C2) metrics, elements are assumed correlated if they have a similar type. In this research, the correlation between elements' inner class is not only based on type similarity but also on the meaning of the element's name. The results confirm that the value of cohesion increases in line with the similarity of meaning.

Identification of Manual Alphabets Based Gestures Using s-EMG for Realizing User Authentication 査読あり

Yamaba H., Usuzaki S., Takatsuka K., Aburada K., Katayama T., Park M., Okazaki N.

Lecture Notes on Data Engineering and Communications Technologies 47 323 - 333 2020年1月

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：Lecture Notes on Data Engineering and Communications Technologies

© 2020, Springer Nature Switzerland AG. At the present time, since mobile devices such as tablet-type PCs and smart phones have penetrated deeply into our daily lives, an authentication method that prevents shoulder surfing attacks comes to be important. We are investigating a new user authentication method for mobile devices that uses surface electromyogram (s-EMG) signals, not screen touching. The s-EMG signals, which are generated by the electrical activity of muscle fibers during contraction, can be detected over the skin surface, and muscle movement can be differentiated by analyzing the s-EMG signals. Taking advantage of the characteristics, we proposed a method that uses a list of gestures as a password in the previous study. In order to realize this method, we have to prepare a sufficient number of gestures that are used to compose passwords. In this paper, we adopted fingerspelling as candidates of such gestures. We introduced manual kana of the Japanese Sign Language syllabary and compared the identification performance of some candidate sets of feature values with adopting support vector machines.

DOI： 10.1007/978-3-030-39746-3_34

Proposal of an Algorithm to Generate VDM++ by Using Words Extracted from the Natural Language Specification 査読あり

Yasuhiro Shigyo, Tetsuro Katayama, Yoshihiro Kita, Hisaaki Yamaba, Kentaro Aburada, Naonobu Okazaki

PROCEEDINGS OF THE 2020 INTERNATIONAL CONFERENCE ON ARTIFICIAL LIFE AND ROBOTICS (ICAROB2020) 763 - 766 2020年1月

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：ALIFE ROBOTICS CO, LTD

The natural language includes ambiguous expressions. VDM is one of methodology on the formal methods to write the specification without ambiguity. Because VDM++ is written by strict grammar, it is difficult to write a VDM++ specification. This research attempts to generate a VDM++ specification automatically from a natural language specification by machine learning. To generate a VDM++ specification, it is necessary to extract words which consist of predicate corresponding to the function and nouns corresponding to variable from the natural language specification. This paper proposes an approach to generate a VDM++ specification from the extracted words list. Identifiers are generated from the extracted words list, and then the VDM++ specification can be generated by converting them into VDM++ grammar.