Statistical Stream Metrics for Software Quality

The Statistical stream metrics developed by us are of unique type (as compared to the existing available metrics) and we propose these metrics as the solution towards software quality. Probably the managers feel they are a bit "techie." We expect that this concise research of the measures has shown that they are practical and pragmatic techniques of assuring quality. The foundation of statistical stream metrics is based upon the principle of FanIn & FanOut or component coupling. Most of the systems consist of components and it is the software performance that these components actually do. The way components are linked or associated together pretty much effect the complexity of a software product. If a component has to do a number of separate tasks it is said to be lacking in "cohesion." Also, systems are highly coupled, if the components within the system communicate data extensively with other components. Systems theory approach talks about that the components which are highly coupled and are less cohesive. These sorts of components with more coupling and less cohesion may be less reliable and difficult to maintain than those components that are loosely coupled and highly cohesive. General Terms Software Quality