Title :
An extension of SEMEST: the online software engineering measurement tool
Author :
Zhang, Shuangshuang ; Wang, Yingxu
Author_Institution :
Dept. of Electr. & Comput. Eng., Calgary Univ., Alta., Canada
Abstract :
Software engineering measurement and metrics are key technologies toward quantitative software engineering. Current software measurement tools are application specific and usually cover only one or a few measures/metrics in software engineering. To address this problem, a comprehensive software engineering measurement expert system tool (SEMEST) was developed. In order to enhance the current version of SEMEST, SEMEST+ is designed by extending the power and performance of its earlier version. SEMEST+ is a Web-based software measurement expert tool to provide a comprehensive set of software measures and metrics in a rigorous way. The extensions of SEMEST are focused on updating the knowledge base and its performance. First, measurements for different application domains and roles of software engineering are added to the knowledge base. This extension enables SEMEST+ to support five domains of software engineering measurement, i.e., goal-, process-, category-, domains-, and roles-oriented measurement, and their analysis. Second, we adopt data mining technologies to analyze large sets of measurement data collected from the software industry. For example, data on defects and productivity may be analyzed and benchmarked by using the tool. Third, recommendations can be derived based on the measurement results in order to improve an organization´s practice and the quality of work products.
Keywords :
data mining; expert systems; software metrics; software process improvement; software tools; application domains; application roles; application specific tools; category-oriented measurement; data mining; domains-oriented measurement; goal-oriented measurement; knowledge base; process-oriented measurement; roles-oriented measurement; software engineering measurement expert system tool; software engineering metrics; software tool; Application software; Computer industry; Current measurement; Data mining; Expert systems; Mining industry; Productivity; Software engineering; Software measurement; Software tools;
Conference_Titel :
Electrical and Computer Engineering, 2004. Canadian Conference on
Print_ISBN :
0-7803-8253-6
DOI :
10.1109/CCECE.2004.1349695