Towards a Framework for Empirical Project Analysis for Software Engineering Models

Data collection and analysis is a central issue in empirical software engineering. This is particularly true for automated gathering of data. Also the empirical evaluation of many research approaches requires the use of combination of data sources from various domains. Capturing and combining the spatial and temporal data from heterogeneous sources is a non-trivial and time-consuming task. In this paper, we propose the Empirical Project Analysis Framework (EPAF), a data analysis framework for a uniform repository as data source for empirical studies based on previous experiences with data mining for empirical studies. The repository contains artifacts from multiple domains thus reducing the effort of integration. In our approach we combine a uniform model repository with operation-based versioning and provide an extensible analyzer framework on top of them. Furthermore we illustrate the use of the proposed framework in three real research projects and present its successful application in their empirical studies.