Article
Computer Science, Artificial Intelligence
Eman Abdullah AlOmar, Anthony Peruma, Mohamed Wiem Mkaouer, Christian Newman, Ali Ouni, Marouane Kessentini
Summary: Recent studies show that developers refactor for a variety of reasons beyond just improving software design and fixing code smells. Developers use a variety of patterns in commit messages to describe their refactoring activities. The distribution of refactoring operations differs between production and test files.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Ashima Kukkar, Umesh Kumar Lilhore, Jaroslav Frnda, Jasminder Kaur Sandhu, Rashmi Prava Das, Nitin Goyal, Arun Kumar, Kamalakanta Muduli, Filip Rezac
Summary: This research article presents a ProRE model based on Ant colony optimization for accurate bug assignment in software engineering. The model utilizes data pre-processing, feature selection, and programmer recommendation stages to allocate bugs to programmers. The proposed model outperforms other existing systems, with an improvement of 4%, 10%, and 12% compared to SVM, C4.5, and NB-based models.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2023)
Article
Chemistry, Multidisciplinary
Seonah Lee, Jaejun Lee, Sungwon Kang, Jongsun Ahn, Heetae Cho
Summary: This paper proposes a code edit recommendation method using a recurrent neural network, which improves the average recommendation accuracy by approximately 5% compared to the state-of-the-art method MI-EA.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Information Systems
Ana Gjorgjevikj, Kostadin Mishev, Ljupcho Antovski, Dimitar Trajanov
Summary: Machine learning methods have brought revolutionary changes to various domains and provided solutions to previously unsolvable problems. With the availability of large amounts of data, powerful processing architectures, and user-friendly software frameworks, machine learning has become a popular and affordable option. However, the development and maintenance of production-level machine learning systems are challenging and require engineering approaches and best practices.
Article
Computer Science, Software Engineering
Zhiyuan Wan, Xin Xia, David Lo, Gail C. Murphy
Summary: The study reveals significant differences between the development of machine learning systems and non-machine-learning systems in various aspects of software engineering and work characteristics. The research involved 14 interviewees and 342 survey respondents from 26 countries across four continents. Additionally, the study highlights future research directions and provides recommendations for practitioners.
IEEE TRANSACTIONS ON SOFTWARE ENGINEERING
(2021)
Article
Computer Science, Software Engineering
Suppawong Tuarob, Noppadol Assavakamhaenghan, Waralee Tanaphantaruk, Ponlakit Suwanworaboon, Saeed-Ul Hassan, Morakot Choetkiertikul
Summary: Building a team in large-scale collaborative software development can be challenging due to the abundance of choices for candidate members. RECAST is an intelligent recommendation system that suggests team configurations based on role requirements, technical skills, and teamwork compatibility. Evaluation results show that RECAST outperforms existing algorithms in team recommendation, improving performance by 646% using exact-match evaluation protocol.
EMPIRICAL SOFTWARE ENGINEERING
(2021)
Article
Computer Science, Theory & Methods
Yanming Yang, Xin Xia, David Lo, John Grundy
Summary: This article introduces the application of deep learning in the field of software engineering research and presents a series of research questions and survey results to summarize and classify different deep learning techniques and analyze aspects such as data processing and optimization algorithms.
ACM COMPUTING SURVEYS
(2022)
Review
Computer Science, Information Systems
Saad Shafiq, Atif Mashkoor, Christoph Mayr-Dorn, Alexander Egyed
Summary: This article examines the adoption of machine learning in the software engineering community to transition software towards highly intelligent and self-learning systems. The study aims to investigate the use of machine learning across various software development life cycle stages, exploring the relationship between different stages and machine learning tools, techniques, and types, to determine if machine learning favors certain stages and techniques.
Article
Chemistry, Analytical
Victor Takashi Hayashi, Wilson Vicente Ruggiero, Julio Cezar Estrella, Artino Quintino Filho, Matheus Ancelmo Pita, Reginaldo Arakaki, Cairo Ribeiro, Bruno Trazzi, Romeo Bulla Jr
Summary: This study proposes a Test Driven Development (TDD) approach based on unsupervised Machine Learning (ML) to automatically assess IoT modules. By monitoring and analyzing IoT devices, device constraint violations can be identified, and recommendations for monitoring frequency configuration can be provided for different firmware versions.
Editorial Material
Multidisciplinary Sciences
Guido C. H. E. de Croon
Summary: An autonomous drone has successfully competed and won against human drone-racing champions, thanks to advanced engineering and an artificial intelligence system that learns predominantly through trial and error.
Editorial Material
Multidisciplinary Sciences
Ying-Lang Wang, Mao-Chih Huang
Summary: Engineers and algorithms have competed in a virtual test to design a step in the process of manufacturing computer chips. Pairing human expertise with computational efficiency proves most cost-effective, but only when the timing is right.
Article
Computer Science, Artificial Intelligence
Guilherme Palumbo, Davide Carneiro, Miguel Guimares, Victor Alves, Paulo Novais
Summary: In recent years, there has been a significant increase in the number of machine learning algorithms and their parameters. This presents both opportunities and challenges in training models. Traditional search-based methods become computationally expensive and time-consuming as datasets grow, especially in data streaming scenarios. This paper proposes a meta-learning approach that can predict performance indicators and recommend the best algorithm/configuration for training models. The proposed approach is up to 130 times faster than a state-of-the-art method and only slightly worse in terms of model quality, making it suitable for scenarios that require regular model updates with shorter training time.
INTERNATIONAL JOURNAL OF NEURAL SYSTEMS
(2023)
Review
Computer Science, Information Systems
Muhammad Khatibsyarbini, Mohd Adham Isa, Dayang N. A. Jawawi, Muhammad Luqman Mohd Shafie, Wan Mohd Nasir Wan-Kadir, Haza Nuzly Abdull Hamed, Muhammad Dhiauddin Mohamed Suffian
Summary: Software quality can be assured through the process of software testing, and one approach is to adopt test case prioritization (TCP) which has multiple techniques with their own strengths and weaknesses. This review paper focuses on exploring machine learning techniques in TCP, with 110 primary studies selected based on research questions. Discussions on various ML techniques in TCP for software testing have shown trends in recent years, with potential for further improvement.
Article
Computer Science, Artificial Intelligence
Khlood Ahmad, Mohamed Abdelrazek, Chetan Arora, Arbind Agrahari Baniya, Muneera Bano, John Grundy
Summary: This paper presents a new framework based on human-centered AI guidelines and a user survey to collect requirements for human-centered AI-based software. The framework is applied to a case study for enhancing the quality of 360 degrees videos for VR users. The approach helps the project team understand the human-centered needs of the project and capture requirements at different stages of the engineering process of AI-based software.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Theory & Methods
Zoe Kotti, Rafaila Galanopoulou, Diomidis Spinellis
Summary: Machine learning techniques improve the effectiveness of software engineering lifecycle activities. We collected, assessed, summarized, and categorized 83 reviews on ML for SE published between 2009 and 2022, covering 6,117 primary studies. ML is most commonly applied in software quality and testing, while human-centered areas pose greater challenges. We propose various research challenges and actions for ML in SE, including further empirical validation and industrial studies, reconsideration of deficient SE methods, documentation and automation of data collection and pipeline processes, reexamination of proprietary data distribution by industrial practitioners, and implementation of incremental ML approaches.
ACM COMPUTING SURVEYS
(2023)