Quick remedy commits and their impact on mining software repositories

Most changes during software maintenance and evolution are not atomic changes, but rather the result of several related changes affecting different parts of the code. It may happen that developers omit needed changes, thus leaving a task partially unfinished, introducing technical debt or injecting bugs. We present a study investigating “quick remedy commits” performed by developers to implement changes omitted in previous commits. With quick remedy commits we refer to commits that (i) quickly follow a commit performed by the same developer, and (ii) aim at remedying issues introduced as the result of code changes omitted in the previous commit (e.g., fix references to code components that have been broken as a consequence of a rename refactoring) or simply improve the previously committed change (e.g., improve the name of a newly introduced variable). Through a manual analysis of 500 quick remedy commits, we define a taxonomy categorizing the types of changes that developers tend to omit. The taxonomy can (i) guide the development of tools aimed at detecting omitted changes and (ii) help researchers in identifying corner cases that must be properly handled. For example, one of the categories in our taxonomy groups the reverted commits, meaning changes that are undone in a subsequent commit. We show that not accounting for such commits when mining software repositories can undermine one’s findings. In particular, our results show that considering completely reverted commits when mining software repositories accounts, on average, for 0.07 and 0.27 noisy data points when dealing with two typical MSR data collection tasks (i.e., bug-fixing commits identification and refactoring operations mining, respectively).

An Integrated Framework for Traceability and Impact Analysis in Requirements Verification of Cyber–Physical Systems

In the field of model-based design of Cyber–Physical Systems (CPS), seamless traceability of the process, from requirements to models to simulation results, is becoming increasingly important. It can be used to support several activities such as variant handling, impact analysis, component reuse, software maintenance and evolution, verification, and validation. Despite the fact that the relevance of traceability in the model-based design of CPSs is well known, current tools that support traceability management are inadequate in practice. The lack of comprehensive whole-lifecycle systems engineering support in a single tool is one of the main causes of such ineffective traceability management, where traceability relationships between artifacts are still manually generated and maintained. This paper aims at presenting an approach and a prototype for automatically generating and maintaining the appropriate traceability links between heterogeneous artifacts ranging from requirement models, through design models, down to simulation and verification results throughout the product life cycle in model-based design of CPSs. A use case study is presented to validate and illustrate the proposed method and prototype.

Classification of Mobile Application Reviews using Word Embedding and Convolutional Neural Network

The app reviews are useful for app developers because they contain valuable information, e.g. bug, feature request, user experience, and rating. This information can be used to better understand user needs and application defects during software maintenance and evolution phase. The increasing number of reviews causes problems in the analysis process for developers. Reviews in textual form are difficult to understand, this is due to the difficulty of considering semantic between sentences. Moreover, manual checking is time-consuming, requires a lot of effort, and costly for manual analysis. Previous research shows that the collection of the review contains non-informative reviews because they do not have valuable information. Non-informative reviews considered as noise and should be eliminated especially for classification process. Moreover, semantic problems between sentences are not considered for the reviews classification. The purpose of this research is to classify user reviews into three classes, i.e. bug, feature request, and non-informative reviews automatically. User reviews are converted into vectors using word embedding to handle the semantic problem. The vectors are used as input into the first classifier that classifies informative and non-informative reviews. The results from the first classifier, that is informative reviews, then reclassified using the second classifier to determine its category, e.g. bug report or feature request. The experiment using 306,849 sentences of reviews crawled from Google Play and F-Droid. The experiment result shows that the proposed model is able to classify mobile application review by produces best accuracy of 0.79, precision of 0.77, recall of 0.87, and F-Measure of 0.81.