To automatically map source code entities to architectural modules with Naive Bayes

The process of mapping a source code entity onto an architectural module is to large degree manual task. Automating this could increase the use static architecture conformance checking methods, such as reflexion modeling, in industry. Current techniques rely on user parameterization and highly cohesive design. A machine learning approach would potentially require less parameters better available information aid automatic mapping. We investigate how classifier can be trained map from modules automatically. This with semantic syntactic dependency extracted descriptions. implemented using multinomial naive Bayes evaluated. perform experiments compare three state-of-the-art functions eight open-source Java systems known ground-truth-mappings. find that outperforms all cases it provides useful baseline for further research area semi-automatic incremental clustering. conclude performs need compared other approaches. Future work includes investigating problematic mappings more diverse set subject systems.