Quality Assessment of ChatGPT Generated Code and their Use by Developers
The release of large language models (LLMs) like ChatGPT has revolutionized software development. Numerous studies are exploring the generated response quality of ChatGPT, the effectiveness of different prompting techniques, and its performance in programming contests, among other aspects. However, there is limited information regarding the practical usage of ChatGPT by software developers. This data mining challenge focuses on DevGPT, a curated dataset of developer-ChatGPT conversations encompassing prompts with ChatGPT’s responses, including code snippets. Our paper leverages this dataset to investigate (RQ1) whether ChatGPT generates Python & Java code with quality issues; (RQ2) whether ChatGPT-generated code is merged into a repository, and, if it does, to what extent developers change them; and (RQ3) what are the main use cases for ChatGPT besides code generation. We found that ChatGPT-generated code suffers from using undefined/unused variables and improper documentation. They are also suffering from improper resources and exception management-related security issues. Our results show that ChatGPT-generated codes are hardly merged, and they are significantly modified before merging. Instead, based on an analysis of developers’ discussions and the developer-ChatGPT chats, we found that developers use this model for every stage of software development and leverage it to learn about new frameworks and development kits.
| Presentation Slides (MSR-Challenge'24.pptx) | 4.63MiB |
Mon 15 AprDisplayed time zone: Lisbon change
14:00 - 15:30 | Mining ChallengeMining Challenge at Almada Negreiros Chair(s): Preetha Chatterjee Drexel University, USA, Fabio Palomba University of Salerno | ||
14:00 5mTalk | ChatGPT Chats Decoded: Uncovering Prompt Patterns for Superior Solutions in Software Development Lifecycle Mining Challenge Liangxuan Wu Huazhong University of Science and Technology, Yanjie Zhao Huazhong University of Science and Technology, Xinyi Hou Huazhong University of Science and Technology, Tianming Liu Monash Univerisity, Haoyu Wang Huazhong University of Science and Technology | ||
14:05 5mTalk | Write me this Code: An Analysis of ChatGPT Quality for Producing Source Code Mining Challenge Konstantinos Moratis Electrical and Computer Engineering Dept., Aristotle University of Thessaloniki, Themistoklis Diamantopoulos Electrical and Computer Engineering Dept, Aristotle University of Thessaloniki, Dimitrios-Nikitas Nastos Electrical and Computer Engineering Dept., Aristotle University of Thessaloniki, Andreas Symeonidis Aristotle University of Thessaloniki Pre-print | ||
14:10 5mTalk | Quality Assessment of ChatGPT Generated Code and their Use by Developers Mining Challenge Mohammed Latif Siddiq University of Notre Dame, Lindsay Roney University of Notre Dame, Jiahao Zhang , Joanna C. S. Santos University of Notre Dame Pre-print Media Attached File Attached | ||
14:15 5mTalk | Analyzing Developer Use of ChatGPT Generated Code in Open Source GitHub Projects Mining Challenge Balreet Grewal University of Alberta, Wentao Lu University of Alberta, Sarah Nadi New York University Abu Dhabi, University of Alberta, Cor-Paul Bezemer University of Alberta Pre-print | ||
14:20 5mTalk | How I Learned to Stop Worrying and Love ChatGPT Mining Challenge Piotr Przymus Nicolaus Copernicus University in Toruń, Poland, Mikołaj Fejzer Nicolaus Copernicus University in Toruń, Jakub Narębski Nicolaus Copernicus University in Toruń, Krzysztof Stencel University of Warsaw Pre-print | ||
14:25 5mTalk | Can ChatGPT Support Developers? An Empirical Evaluation of Large Language Models for Code Generation. Mining Challenge Kailun Jin York University, Chung-Yu Wang York University, Hung Viet Pham York University, Hadi Hemmati York University Pre-print | ||
14:30 5mTalk | The role of library versions in Developer-ChatGPT conversations Mining Challenge Pre-print | ||
14:35 5mTalk | AI Writes, We Analyze: The ChatGPT Python Code Saga Mining Challenge Md Fazle Rabbi Idaho State University, Arifa Islam Champa Idaho State University, Minhaz F. Zibran Idaho State University, Md Rakibul Islam Lamar University DOI Pre-print | ||
14:40 5mTalk | ChatGPT in Action: Analyzing Its Use in Software Development Mining Challenge Arifa Islam Champa Idaho State University, Md Fazle Rabbi Idaho State University, Costain Nachuma Idaho State University, Minhaz F. Zibran Idaho State University DOI Pre-print | ||
14:45 5mTalk | Chatting with AI: Deciphering Developer Conversations with ChatGPT Mining Challenge Suad Mohamed Belmont University, Abdullah Parvin Belmont University, Esteban Parra Belmont University | ||
14:50 5mTalk | Does Generative AI Generate Smells Related to Container Orchestration?: An Exploratory Study with Kubernetes Manifests Mining Challenge Yue Zhang Auburn University, Rachel Meredith Auburn University, Wilson Reaves Auburn University, Julia Coriolano Federal University of Pernambuco, Muhammad Ali Babar School of Computer Science, The University of Adelaide, Akond Rahman Auburn University Pre-print | ||
14:55 5mTalk | On the Taxonomy of Developers' Discussion Topics with ChatGPT Mining Challenge | ||
15:00 5mTalk | How to refactor this code? An exploratory study on developer-ChatGPT refactoring conversations Mining Challenge Eman Abdullah AlOmar Stevens Institute of Technology, AnushKrishna Venkatakrishnan Rochester Institute of Technology, USA, Mohamed Wiem Mkaouer University of Michigan - Flint, Christian Newman , Ali Ouni ETS Montreal, University of Quebec | ||
15:05 5mTalk | Analyzing Developer-ChatGPT Conversations for Software Refactoring: An Exploratory Study Mining Challenge Omkar Sandip Chavan Rochester Institute of Technology, Divya Dilip Hinge Rochester Institute of Technology, Soham Sanjay Deo Rochester Institute of Technology, Yaxuan (Olivia) Wang Rochester Institute of Technology, Mohamed Wiem Mkaouer University of Michigan - Flint | ||
15:10 5mTalk | How Do Software Developers Use ChatGPT? An Exploratory Study on GitHub Pull Requests Mining Challenge Moataz Chouchen ETS, Narjes Bessghaier ETS Montreal, University of Quebec, Mahi Begoug , Ali Ouni ETS Montreal, University of Quebec, Eman Abdullah AlOmar Stevens Institute of Technology, Mohamed Wiem Mkaouer University of Michigan - Flint | ||
15:15 5mTalk | Investigating the Utility of ChatGPT in the Issue Tracking System: An Exploratory Study Mining Challenge Joy Krishan Das University of Saskatchewan, Saikat Mondal University of Saskatchewan, Chanchal K. Roy University of Saskatchewan, Canada Pre-print | ||
15:20 5mTalk | Enhancing User Interaction in ChatGPT: Characterizing and Consolidating Multiple Prompts for Issue Resolution Mining Challenge Saikat Mondal University of Saskatchewan, Suborno Deb Bappon Department of Computer Science, University of Saskatchewan, Canada, Chanchal K. Roy University of Saskatchewan, Canada Pre-print | ||