A Topic Modeling Based Solution for Confirming Software Documentation Quality

Abstract

this paper presents an approach for evaluating and confirming thequality of the external software documentation using topic modeling.Typically, the quality of the external documentation has to mirrorprecisely the organization of the source code. Therefore, the elementsof such documentation should be strongly written, associated, andpresented. In this paper, we use Latent Dirichlet Allocation (LDA) andHELLINGER DISTANCE to compute the similarities between the fragments ofsource code and the external documentation topics. These similaritiesare used in this paper to improve and advance the existing externaldocumentation. Furthermore, these similarities can also be used forevaluating the new documenting process during the evolution phase of thesoftware. The results show that the new approach yields state-of-the-artperformance in evaluating and confirming the existing externaldocumentations quality and superiority.