Is Stack Overflow Obsolete? An Empirical Study of the Characteristics of ChatGPT Answers to Stack Overflow Questions

被引:8
|
作者
Kabir, Samia [1 ]
Udo-Imeh, David N. [1 ]
Kou, Bonan [1 ]
Zhang, Tianyi [1 ]
机构
[1] Purdue Univ, W Lafayette, IN 47907 USA
关键词
stack overflow; q&a; large language model; chatgpt; misinformation;
D O I
10.1145/3613904.3642596
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Q&A platforms have been crucial for the online help-seeking behavior of programmers. However, the recent popularity of ChatGPT is altering this trend. Despite this popularity, no comprehensive study has been conducted to evaluate the characteristics of ChatGPT's answers to programming questions. To bridge the gap, we conducted the first in-depth analysis of ChatGPT answers to 517 programming questions on Stack Overflow and examined the correctness, consistency, comprehensiveness, and conciseness of ChatGPT answers. Furthermore, we conducted a large-scale linguistic analysis, as well as a user study, to understand the characteristics of ChatGPT answers from linguistic and human aspects. Our analysis shows that 52% of ChatGPT answers contain incorrect information and 77% are verbose. Nonetheless, our user study participants still preferred ChatGPT answers 35% of the time due to their comprehensiveness and well-articulated language style. However, they also overlooked the misinformation in the ChatGPT answers 39% of the time. This implies the need to counter misinformation in ChatGPT answers to programming questions and raise awareness of the risks associated with seemingly correct answers.
引用
收藏
页数:17
相关论文
共 50 条
  • [1] An Empirical Study of Obsolete Answers on Stack Overflow
    Zhang, Haoxiang
    Wang, Shaowei
    Chen, Tse-Hsun
    Zou, Ying
    Hassan, Ahmed E.
    IEEE TRANSACTIONS ON SOFTWARE ENGINEERING, 2021, 47 (04) : 850 - 862
  • [2] Comparing emotions in ChatGPT answers and human answers to the coding questions on Stack Overflow
    Fatahi, Somayeh
    Vassileva, Julita
    Roy, Chanchal K.
    FRONTIERS IN ARTIFICIAL INTELLIGENCE, 2024, 7
  • [3] Empirical Study of the Evolution of Python']Python Questions on Stack Overflow
    Syam, Gopika
    Lal, Sangeeta
    Chen, Tao
    E-INFORMATICA SOFTWARE ENGINEERING JOURNAL, 2023, 17 (01)
  • [4] An Empirical Study on the Influence of Social Interactions for the Acceptance of Answers in Stack Overflow
    Zhang, Zhang
    Mao, Xinjun
    Lu, Yao
    Wang, Shangwen
    Lu, Jinyu
    2020 27TH ASIA-PACIFIC SOFTWARE ENGINEERING CONFERENCE (APSEC 2020), 2020, : 425 - 434
  • [5] Haste Makes Waste: An Empirical Study of Fast Answers in Stack Overflow
    Lu, Yao
    Mao, Xinjun
    Zhou, Minghui
    Zhang, Yang
    Wang, Tao
    Li, Zude
    2020 IEEE INTERNATIONAL CONFERENCE ON SOFTWARE MAINTENANCE AND EVOLUTION (ICSME 2020), 2020, : 23 - 34
  • [6] Characterization and Prediction of Questions without Accepted Answers on Stack Overflow
    Yazdaninia, Mohamad
    Lo, David
    Sami, Ashkan
    2021 IEEE/ACM 29TH INTERNATIONAL CONFERENCE ON PROGRAM COMPREHENSION (ICPC 2021), 2021, : 59 - 70
  • [7] Reading Answers on Stack Overflow: Not Enough!
    Zhang, Haoxiang
    Wang, Shaowei
    Chen, Tse-Hsun
    Hassan, Ahmed E.
    IEEE TRANSACTIONS ON SOFTWARE ENGINEERING, 2021, 47 (11) : 2520 - 2533
  • [8] Mining Successful Answers in Stack Overflow
    Calefato, Fabio
    Lanubile, Filippo
    Marasciulo, Maria Concetta
    Novielli, Nicole
    12TH WORKING CONFERENCE ON MINING SOFTWARE REPOSITORIES (MSR 2015), 2015, : 430 - 433
  • [9] An empirical study of question discussions on Stack Overflow
    Wenhan Zhu
    Haoxiang Zhang
    Ahmed E. Hassan
    Michael W. Godfrey
    Empirical Software Engineering, 2022, 27
  • [10] Evaluating Privacy Questions From Stack Overflow: Can ChatGPT Compete?
    Defile, Zack
    Radel, Sean
    Godinez, Joe
    Engstrom, Garrett
    Brucker, Theo
    Young, Kenzie
    Ghanavati, Scpideh
    2023 IEEE 31ST INTERNATIONAL REQUIREMENTS ENGINEERING CONFERENCE WORKSHOPS, REW, 2023, : 239 - 244