Recovering traceability links between code and documentation

Abstract

Abstract-Software system documentation is almost always expressed informally in natural language and free text. Examples include requirement specifications, design documents, manual pages, system development journals, error logs, and related maintenance reports. We propose a method based on information retrieval to recover traceability links between source code and free text documents. A premise of our work is that programmers use meaningful names for program items, such as functions, variables, types, classes, and methods. We believe that the application-domain knowledge that programmers process when writing the code is often captured by the mnemonics for identifiers; therefore, the analysis of these mnemonics can help to associate high-level concepts with program concepts and vice-versa. We apply both a probabilistic and a vector space information retrieval model in two case studies to trace C++ source code onto manual pages and Java code to functional requirements. We compare the results of applying the two models, discuss the benefits and limitations, and describe directions for improvements.

Traceability recovery in RAD software systems

“…Clearly, this required a coding standard to help associating comments with classes and methods. In agreement with previous contributions [4,6], text normalization was fundamental; however, text normalization have to be applied taking into account the identified affecting factors, and thus following the proposed process. By simply applying text normalization, very poor results were obtained.…”

“…While performing the traceability recovery process, we discovered that comments are a valuable source of information, thus on the contrary of previous work [4,6] comments were exploited to recover traceability links. Clearly, this required a coding standard to help associating comments with classes and methods.…”

Source code indexing for automated tracing

“…Requirements-to-source-code traceability, automated through information retrieval (IR) methods, has been proven beneficial in several software engineering tasks such as verification and validation, software reuse, and change impact analysis [1][2][3]. The process consists of three steps: indexing, retrieval and presentation.…”

“…For example, Antoniol et al [1] used tf-idf (a vector space IR model) and unigram approximation (a probabilistic IR model) to establish traceability links between object-oriented code and functional requirements, and Marcus and Maletic [2] exploited latent semantic indexing (LSI) for automatic documentation-to-sourcecode traceability links recovery. The research is primarily concerned with the generation of candidate traceability links with only little attention paid to indexing.…”

Using heuristics to estimate an appropriate number of latent topics in source code analysis

“…While there are many benefits in retaining information about a requirement's life, the lack of automated techniques for generating traceability links can often make recovery a costly process. Antoniol et al first explored the usefulness of using information retrieval techniques to analyse project documentation, including specifications, design documents, logs, and other available related text sources, in an attempt to recover traceability links [20]. The results of their study indicated that these information retrieval techniques afford some ability to recover traceability links in a semi-automatic way.…”