This dissertation, "An Empirical Study of the Use of Conceptual Models for Mutation Testing of Database Application Programs" by Yongjian, Wu, 吳勇堅, was obtained from The University of Hong Kong (Pokfulam, Hong Kong) and is being sold pursuant to Creative Commons: Attribution 3.0 Hong Kong License. The content of this dissertation has not been altered in any way. We have altered the formatting in order to facilitate the ease of printing and reading of the dissertation. All rights not granted by the above license are retained by the author.
Abstract:
Abstract of thesis entitled "An Empirical Study of the Use of Conceptual Models for Mutation Testing of Database Application Programs" Submitted by Yongjian Wu for the degree of Master of Philosophy at The University of Hong Kong in November 2006 Databases are indispensable in current knowledge-based information society. Database applications are those applications that communicate with databases extensively. The ef- fective and efficient manipulation of databases is critical to the quality of database appli- cations. SQL statements are widely used in database applications to communicate with relational databases, the most popular type of databases. It is, however, hard to assure the correctness of SQL statements and prevent the side effects of SQL statements, not to mention the correctness of database applications with many embedded SQL statements. The Extended Entity Relationship model (EER model) is a popular conceptual data model for relational databases. Constraints in the EER model are critical in simulating real-life situations. We investigate the application of mutation testing method based on constraints in the EER model to assure the correct operations of database application pro- grams. This thesis reports an empirical study of effectiveness and efficiency of the method. First of all, we develop three sets of mutants, including two sets of conventional mutants and a set of model-based mutants. Then, we compare the relative mutation scores and the fault detectabilities between conventional mutation testing methods and the model-based mutation testing method. The results show that the model-based mutation testing method is better than conventional mutation testing in detecting faults related to conceptual mod- els. Finally, we investigate the relationship between fault detectability and model-based mutation operators. We find that, if a test suite can kill more types of mutants, it can detect faults more efficiently. We find the same conclusion using different oracles. This thesis also presents our preliminary tools. (259 words)
DOI: 10.5353/th_b3759943
Subjects:
Entity-relationship modeling
Relational databases - Management
Database management - Computer programs
An Empirical Study of the Use of Conceptual Models for Mutation Testing of Database Application Programs
