A Method Of Metamorphic Relationship Sharing Based On Ontology

With the development of testing technology,metamorphic testing has been proposed and gradually become the focus of testing research.Different from traditional testing,metamorphic testing solves testing problems when test predictions cannot be obtained.Metamorphic testing deals with Oracle problems by repeatedly executing input data with a certain relationship to see if the corresponding output results also meet a certain relationship.The success of a metamorphic test is closely related to the selection of metamorphic relationship.In recent years,the application of metamorphic testing is more and more extensive,and has produced a lot of valuable metamorphic relationships(MRs)for different software and programs.However,due to the different descriptions of the metamorphosis relationship and the diversification of the forms of the metamorphosis relationship,the description of the metamorphosis relationship has great variability,so the lack of a standard description mechanism makes the metamorphosis relationship be shared and reused.In this paper,ontology is used to model the metamorphic relation,and a shared template for metamorphic relation description is abstracting.Finally,a metamorphic relation query system based on ontology semantic web is designed and completed to realize the query and sharing of metamorphic relation.As a research hotspot in the field of knowledge engineering,ontology has strong semantic description ability and reasoning ability,and is easy to expand and realize knowledge sharing and reuse at the semantic level.Traditional queries are mostly based on keywords,so the real needs of users cannot be understood from the semantic level.Using semantic web query can improve user satisfaction.In terms of semantic query,this paper chooses database-based semantic retrieval,which is simple and easy to use,and the search results are more accurate.It mainly achieves semantic expansion in two aspects: one is to convert the keywords entered by users into keywords in the relational database;Second,the use of correlation to enrich the query results.For the former,the similarity calculation algorithm is used torealize it.According to the requirements of the application field of the transformation relationship sharing,it not only involves the similarity calculation of English words but also the similarity calculation of English phrases.Therefore,on the basis of the existing similarity algorithm,this paper proposes an improved similarity calculation method based on Wordnet.For the latter,this paper makes use of upper and lower bit relation to conduct semantic association query.In general,the system extends the user's query input and calculates the similarity,and finally realizes the user's intelligent retrieval by combining with Jena framework.