UML diagrams are a recognized standard modelling language for representing design of software systems.For academic research, large cases containing UML diagrams are needed.One of the challenges in collecting such datasets is automatically determining whether an image is a UML diagram or not and what type of UML diagram an image contains.In this study, we collected UML diagrams from open datasets and manually labeled them into four types of UML diagrams (i.e., class diagrams, activity diagrams, sequence diagrams, and use case diagrams) and non-UML images.We evaluated the performance of five popular neural network architectures using transfer learning on the dataset of 3231 images that contains 700 class diagrams, 454 activity diagrams, 651 use case diagrams, 706 sequence diagrams, and 720 non-UML images, respectively.We also proposed our neural network architecture for multiclass classification of UML diagrams.The experiment results show that our proposed neural network architecture achieved the best performance amongst the algorithms we evaluated with an accuracy of 98.65%, a precision of 96.76%, a recall of 96.48%, and an F1-score of 96.62%.Moreover, among the neural network architectures that we have evaluated, our proposed architecture has the least parameters (around 2.4 millions) and spends the least time per image (0.0135 seconds per image using GPU) for classifying UML diagrams.

Unified Modeling Language (UML) diagrams are a recognized standard modeling language for representing design of software systems. For academic research, large cases containing UML diagrams are needed. One of the challenges in collecting such datasets is automatically determining whether an image is a UML diagram or not and what type of UML diagram an image contains. In this work, we collected UML diagrams from open datasets and manually labeled them into 10 types of UML diagrams (i.e. class diagrams, activity diagrams, use case diagrams, sequence diagrams, communication diagrams, component diagrams, deployment diagrams, object diagrams, package diagrams, and state machine diagrams) and non-UML images. We evaluated the performance of seven popular neural network architectures using transfer learning on the dataset of 4706 images, including 700 class diagrams, 454 activity diagrams, 651 use case diagrams, 706 sequence diagrams, 204 communication diagrams, 208 component diagrams, 287 deployment diagrams, 207 object diagrams, 246 package diagrams, 323 state machine diagrams, and 720 non-UML images, respectively. We also proposed our neural network architecture for multiclass classification of UML diagrams. The experiment results show that Xception achieved the best performance amongst the algorithms we evaluated with a precision of 93.03%, a recall of 92.44%, and an F1-score of 92.73%. Moreover, it is possible to develop small and almost the same efficient neural network architectures, that our proposed architecture has the least parameters (around 2.4 millions) and spends the least time per image (0.0135[Formula: see text]s per image using graphics processing unit) for classifying UML diagrams with a precision of 91.25%, a recall of 90.34%, and an F1-score of 90.79%.

Background: Code smells indicate potential design or implementation problems that may have a negative impact on programs. Similar to other software artefacts, developers use Stack Overflow (SO) to ask questions about code smells. However, given the high number of questions asked on the platform, and the limitations of the default tagging system, it takes significant effort to extract knowledge about code smells by means of manual approaches. Aim: We utilized supervised machine learning techniques to automatically identify code-smell discussions from SO posts. Method: We conducted an experiment using a manually labeled dataset that contains 3000 code-smell and 3000 non-code-smell posts to evaluate the performance of different classifiers when automatically identifying code smell discussions. Results: Our results show that Logistic Regression (LR) with parameter C=20 (inverse of regularization strength) and Bag of Words (BoW) feature extraction technique achieved the best performance amongst the algorithms we evaluated with a precision of 0.978, a recall of 0.965, and an F1-score of 0.971. Conclusion: Our results show that machine learning approach can effectively locate code-smell posts even if posts' title and/or tags cannot be of help. The technique can be used to extract code smell discussions from other textual artefacts (e.g., code reviews), and promisingly to extract SO discussions of other topics.