A context aware framework for product based software certification

Abstract

<p>p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 11.0px Times}</p> <p>Software certification is becoming a reasonable expectation from the ever growing number of software users. The process of software certification could be described as a process in which an auditing body ensures that the product conforms to certain requirements. The certification models which were analyzed included product based, component based and model based. With our results and findings we have developed a proof of concept context aware framework for product based software certification. The proposed product based software certification process is structured on component based certification principals, while the general core of the certification process is composed from known software certification models.</p> <p>The framework was developed within an OSGi (Open Service Gateway initiative) environment which is being managed by a collection of automation scripts. The certification models which we reviewed did not represent their knowledge formally and did not have any mechanisms to derive indirect knowledge. To address this issue, we have developed an upper ontology to formally model higher level concepts for the certification, and described a general metric for assigning consistent weights to ontology classes. The framework provides a dynamic environment for the certification process by integrating development and certification domains with the help of ontology. Its main objective is to allow the certification process to be able to adjust to ever changing certification demands and extend more easily into different domains.</p> <p>The developed ontology can maintain many properties and attributes, but for some of these the measuring mechanism are unknown. Therefore, we have described the process on how to derive software metrics for measurable and subjective attributes which can be used to evaluate product, processes and resources. In conclusion, we have outlined some areas for future research.</p>