S-CASE Blog | Importing UML Activity Diagrams

One of the main goals of the S-CASE project is to extract requirements from formal models such as UML diagrams. In order to achieve this goal, a robust mechanism for the transformation of formal UML models into the S-CASE ontology is needed.

CERTH has recently focused on applying image recognition techniques and algorithms for the detection and recognition of UML Activity Diagrams representing workflows of stepwise activities and actions.
This process imposes research challenges such as the detection and recognition of the UML glyphs, as well as the recognition and discrimination of the text contained in the diagram. The prototype tool that has been developed analyses UML Activity Diagrams of bmp, jpg and png image formats and produces the XMI scheme representing the original Activity Diagram.

The produced XMI scheme is compliant with the Eclipse Papyrus open source platform, thus the S-CASE user may apply corrections in case of recognition errors through a graphical interface. Future work will be focused the production of a tool that effectively recognises UML Use Case diagrams. These two tools will be wrapped as a single Eclipse plugin and will be available to the Eclipse modelling community.

S-CASE Blog | Web Service Quality Representation

On the blog today we explore the S-CASE approach to Web Service Quality Representation courtesy of Davide Tosi from the Università degli Studi dell’Insubria.

We recently presented the S-CASE Approach of Web Service Quality Representation at the International Conference on Service-Oriented System Engineering (SOSE 2015) in San Francisco Bay, USA. This was a great opportunity to present the S-CASE platform and to focus on the tool we developed to semantically annotate Web Services (WS) with quality aspects.

Quality is an important issue for WS-based software systems, like for any other software system. Quality is intrinsically a multi-faceted concept and is usually defined in terms of several quality attributes, quantified via a number of measures. In addition, relations may exist among the different quality attributes and among their measures. A quality attribute may positively or negatively influence another quality attribute. Each stakeholder needs to be able to evaluate the quality attributes that are of interest in his or her applications for his or her goals. For instance, WS users can evaluate specific quality attributes to decide whether to use a WS. WS providers can use quality attributes to set quality goals for a WS so that Service Level Agreements are satisfied.

However, a limited number of qualities of individual and composite WSs have been studied so far, e.g., reliability, response time, and availability. In addition, none of the available specifications are able to describe the relations among the different quality attributes. In the context of the S-CASE project, we developed a comprehensive approach to representing quality attributes in WS. While compatible with existing specifications defined by OASIS, our representation (we called WSQR: Web Service Quality Representation) adds several important elements that take into account a number of important developments in the Software Measurement research and practice. Our representation proposal therefore enriches these specifications and provides a comprehensive reference model that can be used to syntactically specify internal and external quality attributes and their measures, along with their characteristics and relations. Our proposal is also compatible with the theoretical aspects required by the general literature on Measurement and with the ISO/IEC25010 quality standard.

To read more about Web Service Quality Representation you can read my conference paper from SOSE 2015 right here.

– Davide