Accepted papers

Technical papers

Comparing Within- and Cross-Project Machine Learning Algorithms for Code Smell Detection

Manuel De Stefano (University of Salerno), Fabiano Pecorelli (University of Salerno), Fabio Palomba (University of Salerno),
Andrea De Lucia (University of Salerno)

Abstract: Code smells represent a well-known problem in software engineering, since they are known cause of loss of comprehensibility and maintainability. The most recent efforts in devising automatic machine learning-based code smell detection techniques have achieved unsatisfying results so far. This could be explained by the fact that all these approaches follow a within-project classification, i.e., training and test data are taken from the same source project, which combined with the unbalanced nature of the problem, produces datasets with a very low number of instances belonging to the minority class (i.e., smelly instances). In this paper, we propose a cross-project machine learning approach and comparing its performance with a within-project alternative. The core idea is to use transfer learning to increase the overall number of smelly instances in the training datasets. Our results have shown that cross-project classification provides very similar performance with respect to within-project. Despite this finding does not yet provide a step forward in increasing the performance of ML techniques for code smell detection, it sets the basis for further investigations.

Unsupervised Learning of General-Purpose Embeddings for Code Changes

Mikhail Pravilov (Higher School of Economics), Egor Bogomolov (JetBrains Research, Higher School of Economics), Yaroslav Golubev (JetBrains Research), Timofey Bryksin (JetBrains Research, Higher School of Economics)

Abstract: A lot of problems in the field of software engineering — bug fixing, commit message generation, etc. — require analyzing not only the code itself but specifically code changes. Applying machine learning models to these tasks requires us to create numerical representations of the changes, i.e. embeddings. Recent studies demonstrate that the best way to obtain these embeddings is to pre-train a deep neural network in an unsupervised manner on a large volume of unlabeled data and then further fine-tune it for a specific task. In this work, we propose an approach for obtaining such embeddings of code changes during pre-training and evaluate them on two different downstream tasks — applying changes to code and commit message generation. The pre-training consists of the model learning to apply the given change (an edit sequence) to the code in a correct way, and therefore requires only the code change itself. To increase the quality of the obtained embeddings, we only consider the changed tokens in the edit sequence. In the task of applying code changes, our model outperforms the model that uses full edit sequences by 5.9 percentage points in accuracy. As for the commit message generation, our model demonstrated the same results as supervised models trained for this specific task, which indicates that it can encode code changes well and can be improved in the future by pre-training on a larger dataset of easily gathered code changes.

VaryMinions: Leveraging RNNs to Identify Variants in Event Logs

Sophie Fortz (Université de Namur), Paul Temple (Université de Namur), Xavier Devroey (Delft University of Technology),
Patrick Heymans (Université de Namur), Gilles Perrouin (Université de Namur)

Abstract: Business processes have to manage variability in their execution, e.g., to deliver the correct building permit in different municipalities. This variability is visible in event logs, where sequence of events are shared by the core process (building permit authorisation) but may also be specific to each municipality. To rationalise resources (e.g., derive a configurable business process capturing all municipalities’ permit variants) or to debug anomalous behaviours, it is mandatory to identify to which variant a given trace belongs to. This paper supports this task by training Long Short Term Memory (LSTMs) and Gated Recurrent Units (GRUs) algorithms on two datasets: a configurable municipality and a travel expenses workflow. We demonstrate that variability can be identified accurately (> 87%) and discuss challenges of learning highly entangled variants.

Toward Static Test Flakiness Prediction: A Feasibility Study

Valeria Pontillo (University of Salerno), Fabio Palomba (University of Salerno), Filomena Ferrucci (University of Salerno)

Abstract: Flaky tests are tests that exhibit both a passing and failing behavior when run against the same code. While the research community has attempted to define automated approaches for detecting and addressing test flakiness, most of them suffer from scalability issues and uncertainty as they require test cases to be run multiple times. This limitation has been recently targeted by means of machine learning solutions that could predict the flakiness of tests using a set of both static and dynamic metrics that would avoid the re-execution of tests. Recognizing the effort spent so far, this paper poses the first steps toward an orthogonal view of the problem, namely the classification of flaky tests using only statically computable software metrics. We propose a feasibility study on 72 projects of the iDFlakies dataset, and investigate the differences between flaky and non-flaky tests in terms of 25 test and production code metrics and smells. First, we statistically assess those differences. Second, we build a logistic regression model to verify the extent to which the differences observed are still significant when the metrics are considered together. The results show a relation between test flakiness and a number of test and production code factors, indicating the possibility to build classification approaches that exploit those factors to predict test flakiness.

Building a Bot for Automatic Expert Retrieval on Discord

Ignacio Nuñez Norambuena (University of Chile), Alexandre Bergel (University of Chile)

Abstract: It is common for software practitioners to look for experts on on-line chat platforms, such as Discord. However, finding them is a complex activity that requires a deep knowledge of the open source community. As a consequence, newcomers and casual participants may not be able to adequately find experts willing to discuss a particular topic. Our paper describes a bot that provides a ranked list of Discord users that are experts in a particular set of topics. Our bot uses simple heuristics to model expertise, such as a word occurrence table and word embeddings. Our bot shows that at least half of the retrieved users are indeed experts.

Metrics Selection for Load Monitoring of Service-Oriented System

Francesco Lomio (Tampere University), Sampsa Jurvansuu (Tampere University), Davide Taibi (Tampere University)

Abstract: Background. Complex software systems produce a large amount of data depicting their internal state and activities. The data can be monitored to make estimations and predictions of the status of the system, helping taking preventative actions in case of impending malfunctions and failures. However, a complex system may reveal thousands of internal metrics, which makes it a non-trivial task to decide which metrics are the most important to monitor. Objective. In this work we aim at finding a subset of metrics to collect and analyse for the monitoring of the load in a Service- oriented system. Method. We use a performance test bench tool to generate load of different intensities on the target system, which is a specific service- oriented application platform. The numeric metrics data collected from the system is combined with the load intensity at each moment. The combined data is used to analyse which metrics are best at estimating the load of the system. By using a regression analysis it was possible to rank the metrics by their ability to measure the load of the system. Results. The results show that (1) the use of machine learning regressor allows to correctly measure the load of a system-oriented system, and (2) the most important metrics are related to network traffic and request counts, as well as memory usage and disk activity. Conclusion. The results help with the designs of efficient monitoring tool. In addition, further investigation should be focused on exploring more precise machine learning model to further improve the metric selection process.


The Impact of Release-based Validation on Software Vulnerability Prediction Models

Giulia Sellitto (University of Salerno), Filomena Ferrucci (University of Salerno)

Abstract: Software vulnerability prediction models represent a promising approach in security defect analysis, since they allow to focus testing and improve software quality. Several studies have proposed Machine Learning approaches which demonstrated satisfying performance when evaluated using K-fold Cross-validation. However, recent research demonstrated that, when applying a release-based validation strategy instead, the performance declined. We want to investigate this issue, by conducting a comparative study on different models and dataset. We analyze the impact of using a relase-based validation approach on vulnerability prediction models formerly evaluated using cross-validation. We rely on an existing dataset to evaluate two prediction models that exploit code metrics and textual features, respectively. We confirm that the release-based validation approach leads to generally lower performance, high- lighting that further research would be needed to make vulnerability prediction models more effective.

On the Limitations of Bots for Software Engineering

Sami Hadouaj (INSAT, Tunisia), Fabio Palomba (University of Salerno)

Abstract: Software engineering projects are typically developed by multiple developers that collaboratively work through the use of a version control system. The collaborative nature of software development has given rise to the use of automated mechanisms, called bots, that recommend source code quality-related improvements while a new pull request is open on the repository. These bots use a combination of multiple artificial intelligence approaches, including natural language processing and deep learning. In this presentation abstract, we aim at discussing the current inner-working of these bots as well as the existing limitations. The final goal of our research is to devise a list of improvements that should be made to make those bots more useful to developers.

Evidence and Machine Learning based Task Allocation: a Combined Approach

Stefano Lambiase (University of Salerno), Fabiano Pecorelli (University of Salerno), Fabio Palomba (University of Salerno), Andrea De Lucia (University of Salerno),
Filomena Ferrucci (University of Salerno), Raffaela Mirandola (Politecnico di Milano), Damian Andrew Tamburri (Jheronimus Academy of Data Science)

Abstract: Software construction is nowadays regulated by means of task issuing and allocations—think of issue-tracking systems and the task management therein—, however, task allocation is mostly left to human interpretation with little or no evidence-based recommendation as to which developer (or even online source) might better fit a specific task description. In this presentation abstract we show a software task allocation technique that is evidence-based, namely, it uses a metrics-based approach to figure out which developer, with which skills, better fits a specific task description. Moreover, we propose the application of machine learning techniques to improve the allocation strategy effectiveness over time. We aim to start a fruitful discussion on the application of such strategies to the task allocation activity, which is one of the key factors for the software projects’ success.