• Journal of Software Engineering Society
• /
• v.28 no.1
• /
• pp.23-28
• /
• 2019

Recently, in order to effectively test deep neural network model for image processing application, researches have actively conducted to automatically generate data in corner-case that is not correctly predicted by the model. This paper proposes test data generation method that selects arbitrary words from input of system and transforms them into synonyms in order to test the bug reporter automatic assignment system based on sentence classification deep neural network model. In addition, we compare and evaluate the case of using proposed test data generation and the case of using existing difference-inducing test data generations based on various neuron coverages.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.12 no.2
• /
• pp.189-197
• /
• 2012

It is usual for test data generation to be performed using either high-level specifications or source codes written in high-level programming languages. In certain circumstances, however, such information is not always available. This paper presents a technique that generates test data based on executable object codes. The proposed technique makes use of a very simple function minimization technique without sophisticated object code analysis and produces test data dynamically. We have conducted a simple experiment to evaluate the effectiveness of the proposed test data generation technique with a triangle classification program to show that branch coverage can be easily achieved.

• Journal of Software Engineering Society
• /
• v.24 no.2
• /
• pp.35-45
• /
• 2011

Simulated annealing has been studied a long times. And it is one of the effective techniques for test data generation. But basic SA methods showed bad performance because of neighborhood selection strategies in the case of large input domain. To overcome this limitation, we propose new neighborhood selection approach, Branch Distance. We performs case studies based on the proposed approach to evaluate it's performance and to compare it whit basic SA and Random test generation. The results of the case studies appear that proposed approach show better performance than the other approach.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.04a
• /
• pp.598-601
• /
• 2017

소프트웨어 테스트는 시스템의 신뢰도를 판단하는 중요한 작업이지만, 많은 노력과 비용을 필요로 한다. 모델 기반 테스트는 이러한 비용을 줄이기 위한 방안으로써 제안되었다. 정형적 모델로부터 시스템의 실행 가능한 경로를 파악하고, 각 경로마다 입력 값을 생성하여 테스트를 수행한다. 이 때, 적절한 입력 값을 찾기 위해 메타-휴리스틱 기법을 사용하는데, 기존의 알고리즘은 목적 경로와 관련이 없는 변수까지 구분없이 고려하기 때문에 시스템이 복잡할수록 불필요한 연산이 많아지는 문제가 있다. 본 논문은 슬라이싱 기법과 우선순위 정책을 적용한 테스트 데이터 자동 생성 기법을 제안하며, 실험을 통해 기존의 방법보다 효과적으로 테스트 데이터를 생성함을 보인다.

• Journal of KIISE:Software and Applications
• /
• v.37 no.12
• /
• pp.882-891
• /
• 2010

Many automated dynamic test data generation technique have been proposed. The techniques evaluate fitness of test data through executing instrumented Software Under Test (SUT) and then generate new test data based on evaluated fitness values and optimization algorithms. Previous researches and experiments have been showed that these techniques generate effective test data. However, optimization algorithms in these techniques incur much time to generate test data, which results in huge test case generation cost. In this paper, we propose a technique for reducing the time of evaluating a fitness of test data among steps of dynamic test data generation methods. We introduce the concept of Fitness Evaluation Program (FEP), derived from a path constraint of SUT. We suggest a test data generation method based on FEP and implement a test generation tool, named ConGA. We also apply ConGA to generate test cases for C programs, and evaluate efficiency of the FEP-based test case generation technique. The experiments show that the proposed technique reduces 20% of test data generation time on average.

• Journal of KIISE:Software and Applications
• /
• v.36 no.10
• /
• pp.793-799
• /
• 2009

Esterel is an imperative synchronous language that is well-adopted to specify reactive systems. Programmers sometimes want simple validations that can be applied while the system is under development. Since a reactive system reacts to environment changes, a test data is a sequence of input events. Generating proper test data by hand is complex and error-prone. Although several test data generators exist, they are hard to learn and use. Mostly, system designers need test data to reach a specific status of a target program. In this paper, we develop a test data generator to generate test input sequences for debugging Esterel programs. Our tool is focused on easy usage; users can describe test data properties with simple specifications. We show a case study in which the test data generator is used for a practical development process.

• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.3
• /
• pp.268-276
• /
• 2002

In this paper we suggest a method to produce automatically teat data using UML development diagrams, and analytically describe the application of a tool, Automatic UML-based Test Data Generation (AUTEG) developed using XML technology, to the examples of insurance system. Our AUTEG automatically generates test diagrams that enable to detect errors existing at the interface area between modules composing the whole system, along with test data by applying the existing white-box test technique to the test diagram. Our AUTEG can be applied to the integration test as well as the system test and using the tool, users may make the unit modules of the integration test into several groups.

• The KIPS Transactions:PartD
• /
• v.16D no.6
• /
• pp.911-920
• /
• 2009

Airbone software must comply the DO-178B standard in order to be certified by the FAA. The standard requires the unit testing of safety-critical software to meet the coverage criterion called MC/DC(Modified Condition/Decision Coverage). Although MC/DC is known to be effective in finding errors related to safety, it is also true that generating test cases which satisfy the MC/DC criterion is not easy. This paper presents a tool named MD-SAT which generates MC/DC test cases with SAT(SATisfiability) technology. It can be employed for generating diverse test cases in tools implementing various testing techniques including decision table based test, cause-effect graphing, and state-based test.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10d
• /
• pp.106-108
• /
• 2002

신뢰성있는 소프트웨어의 개발을 위해 테스트의 중요성은 매우 크다. 특히, 최근에 점진적이고 반복적인 소프트웨어 개발 방법론이 각광을 받으면서 소프트웨어의 잦은 변경에 따른 회귀 테스트의 중요성이 점점 커지고 있다. 이에 따라 단위 데스트의 자동화에 대한 연구가 활발히 진행되고 있다. JUnit은 자바 클래스의 단위 레벨 테스팅을 도와 주는 테스트 지원 프레임워크이다. 또한, JTestCase는 테스트 데이터와 테스트 코드를 분리함으로써, 데이터 중심 테스팅(data-driven testing)을 지원하기 위해 개발된 JUnit 확장 프레임워크이다. 본 논문에서는, 이 두 개의 테스트 프레임워크와 자바 리플렉션 API를 이용하여, 하나의 클래스 파일을 읽어 들여 XML 형태의 테스트 데이터 파일과 테스트 드라이버 코드를 자동생성하는 도구를 제안한다. 그리고, 구체적인 예를 통해 본 논문에서 제안하는 도구의 유용성을 보여준다. 본 논문의 데스트 도구는 회귀 단위 테스트에 필요한 노력을 줄여주고, 자바 클래스 단위 테스트를 지원하는 도구 개발의 기반 기술을 제공하며, 궁극적으로 소프트웨어 개발의 생산성을 향상시켜 준다.

• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.8
• /
• pp.401-410
• /
• 2017

This paper proposes a method to generate test cases in an automatic manner, based on checklist containing test cases used for testing embedded systems. In general, the items to be tested are defined in a checklist. However, most test case generation strategies recommend to test a system with not only the defined test items but also various mutated test conditions. The proposed method parses checklist in Korean file and figures out the system inputs and outputs, and operation information. With the found information and the user defined test case generation strategy, the test cases are automatically generated. With the proposed method, the errors introduced during manual test case generation may be reduced and various test cases not defined in checklist can be generated. The proposed method is implemented and the experiment is performed with the checklist for an medical embedded system. The feasibility of the proposed method is shown through the test cases generated from the checklist. The test cases are adequate to the coverages and their statistics are correct.