题目：蜕变测试：一种简单的针对非可测试程序测试方向（Metamorphic Testing: A Simple Methodology for Testing Non-Testable Programs）
主讲简介：Tsong Yueh Chen，Swinburne University of Technology, Australia.教授。研究专长：软件工程、软件测试。Tsong Yueh Chen (陳宗岳) received his BSc and MPhil from The University of Hong Kong, MSc and DIC from the Imperial College of The London University, and PhD from The University of Melbourne. He is currently a Chair Professor of Software Engineering at Swinburne University of Technology, Australia. Prior to joining Swinburne, he taught at The University of Hong Kong and The University of Melbourne. His main research interests include software testing and debugging. Professor Chen is the inventor of the techniques of adaptive random testing (ART) and metamorphic testing (MT). He has published more than 100 papers in the top journals and conferences of computer science area including IEEE Trans. on Computer, IEEE Trans. on Software engineering, ACM Trans. on Software Engineering and Methodology, ACM Computing Surveys, etc.
主讲内容：A program is said to be non-testable if the correctness of its outputs cannot be verified. Since non-testable programs are frequently encountered, a challenging problem has been how to test them. In this seminar, we present a simple, yet effective, methodology, metamorphic testing, which alleviates this challenge. Although it may not be possible to verify the correctness of any particular input of a non-testable program, we may be able to identify relations that should hold between certain inputs and relevant outputs, and through the verification of these relations (which are known as the metamorphic relations), we are able to test the non-testable programs. Although the concept of metamorphic testing is very simple, its scope of applicability is quite extensive, and not restricted only to software testing. This seminar will cover several aspects of metamorphic testing, including the application of metamorphic testing to various types of non-testable programs, the formation of a theory for metamorphic testing, as well as the integration of metamorphic testing with other techniques.