Improved searchability of bug reports using content-based labeling with machine learning of sentences

Yuki Noyori, Hironori Washizaki, Yoshiaki Fukazawa, Hideyuki Kanuka, Keishi Ooshima, Ryosuke Tsuchiya

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    Most stakeholders refer to past bug reports when they encounter a problem since bug reports contain useful information. However, searching for specific content is difficult because there are many bug reports. The desired content depends on the viewpoint of the stakeholder. A full text search includes unwanted content, which is costly. Although this problem has been previously noted, a solution has yet to be proposed. Herein we propose Content-based Labeling Method as a solution. This method organizes information in a bug report by labeling each sentence based on its contents, allowing stakeholders’ viewpoints to be considered. We evaluate the improvement in searchability. The Content-based Labeling Method improves the searchability according to the F-measure and precision of the experimental results.

    Original languageEnglish
    Title of host publicationKnowledge-Based Software Engineering
    Subtitle of host publication2018 - Proceedings of the 12th Joint Conference on Knowledge-Based Software Engineering, JCKBSE 2018
    EditorsFumihiro Kumeno, Konstantinos Oikonomou, Maria Virvou
    PublisherSpringer Science and Business Media Deutschland GmbH
    Pages75-85
    Number of pages11
    ISBN (Print)9783319976785
    DOIs
    Publication statusPublished - 2019 Jan 1
    Event12th Joint Conference on Knowledge-Based Software Engineering, JCKBSE 2018 - Corfu, Greece
    Duration: 2018 Aug 272018 Aug 30

    Publication series

    NameSmart Innovation, Systems and Technologies
    Volume108
    ISSN (Print)2190-3018
    ISSN (Electronic)2190-3026

    Other

    Other12th Joint Conference on Knowledge-Based Software Engineering, JCKBSE 2018
    CountryGreece
    CityCorfu
    Period18/8/2718/8/30

    Fingerprint

    Labeling
    Learning systems
    Machine learning
    Stakeholders

    Keywords

    • Bug report
    • Labeling
    • Machine learning
    • Searchability

    ASJC Scopus subject areas

    • Decision Sciences(all)
    • Computer Science(all)

    Cite this

    Noyori, Y., Washizaki, H., Fukazawa, Y., Kanuka, H., Ooshima, K., & Tsuchiya, R. (2019). Improved searchability of bug reports using content-based labeling with machine learning of sentences. In F. Kumeno, K. Oikonomou, & M. Virvou (Eds.), Knowledge-Based Software Engineering: 2018 - Proceedings of the 12th Joint Conference on Knowledge-Based Software Engineering, JCKBSE 2018 (pp. 75-85). (Smart Innovation, Systems and Technologies; Vol. 108). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-319-97679-2_8

    Improved searchability of bug reports using content-based labeling with machine learning of sentences. / Noyori, Yuki; Washizaki, Hironori; Fukazawa, Yoshiaki; Kanuka, Hideyuki; Ooshima, Keishi; Tsuchiya, Ryosuke.

    Knowledge-Based Software Engineering: 2018 - Proceedings of the 12th Joint Conference on Knowledge-Based Software Engineering, JCKBSE 2018. ed. / Fumihiro Kumeno; Konstantinos Oikonomou; Maria Virvou. Springer Science and Business Media Deutschland GmbH, 2019. p. 75-85 (Smart Innovation, Systems and Technologies; Vol. 108).

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Noyori, Y, Washizaki, H, Fukazawa, Y, Kanuka, H, Ooshima, K & Tsuchiya, R 2019, Improved searchability of bug reports using content-based labeling with machine learning of sentences. in F Kumeno, K Oikonomou & M Virvou (eds), Knowledge-Based Software Engineering: 2018 - Proceedings of the 12th Joint Conference on Knowledge-Based Software Engineering, JCKBSE 2018. Smart Innovation, Systems and Technologies, vol. 108, Springer Science and Business Media Deutschland GmbH, pp. 75-85, 12th Joint Conference on Knowledge-Based Software Engineering, JCKBSE 2018, Corfu, Greece, 18/8/27. https://doi.org/10.1007/978-3-319-97679-2_8
    Noyori Y, Washizaki H, Fukazawa Y, Kanuka H, Ooshima K, Tsuchiya R. Improved searchability of bug reports using content-based labeling with machine learning of sentences. In Kumeno F, Oikonomou K, Virvou M, editors, Knowledge-Based Software Engineering: 2018 - Proceedings of the 12th Joint Conference on Knowledge-Based Software Engineering, JCKBSE 2018. Springer Science and Business Media Deutschland GmbH. 2019. p. 75-85. (Smart Innovation, Systems and Technologies). https://doi.org/10.1007/978-3-319-97679-2_8
    Noyori, Yuki ; Washizaki, Hironori ; Fukazawa, Yoshiaki ; Kanuka, Hideyuki ; Ooshima, Keishi ; Tsuchiya, Ryosuke. / Improved searchability of bug reports using content-based labeling with machine learning of sentences. Knowledge-Based Software Engineering: 2018 - Proceedings of the 12th Joint Conference on Knowledge-Based Software Engineering, JCKBSE 2018. editor / Fumihiro Kumeno ; Konstantinos Oikonomou ; Maria Virvou. Springer Science and Business Media Deutschland GmbH, 2019. pp. 75-85 (Smart Innovation, Systems and Technologies).
    @inproceedings{a1501ec47fd0463ca93e946ad7f15383,
    title = "Improved searchability of bug reports using content-based labeling with machine learning of sentences",
    abstract = "Most stakeholders refer to past bug reports when they encounter a problem since bug reports contain useful information. However, searching for specific content is difficult because there are many bug reports. The desired content depends on the viewpoint of the stakeholder. A full text search includes unwanted content, which is costly. Although this problem has been previously noted, a solution has yet to be proposed. Herein we propose Content-based Labeling Method as a solution. This method organizes information in a bug report by labeling each sentence based on its contents, allowing stakeholders’ viewpoints to be considered. We evaluate the improvement in searchability. The Content-based Labeling Method improves the searchability according to the F-measure and precision of the experimental results.",
    keywords = "Bug report, Labeling, Machine learning, Searchability",
    author = "Yuki Noyori and Hironori Washizaki and Yoshiaki Fukazawa and Hideyuki Kanuka and Keishi Ooshima and Ryosuke Tsuchiya",
    year = "2019",
    month = "1",
    day = "1",
    doi = "10.1007/978-3-319-97679-2_8",
    language = "English",
    isbn = "9783319976785",
    series = "Smart Innovation, Systems and Technologies",
    publisher = "Springer Science and Business Media Deutschland GmbH",
    pages = "75--85",
    editor = "Fumihiro Kumeno and Konstantinos Oikonomou and Maria Virvou",
    booktitle = "Knowledge-Based Software Engineering",

    }

    TY - GEN

    T1 - Improved searchability of bug reports using content-based labeling with machine learning of sentences

    AU - Noyori, Yuki

    AU - Washizaki, Hironori

    AU - Fukazawa, Yoshiaki

    AU - Kanuka, Hideyuki

    AU - Ooshima, Keishi

    AU - Tsuchiya, Ryosuke

    PY - 2019/1/1

    Y1 - 2019/1/1

    N2 - Most stakeholders refer to past bug reports when they encounter a problem since bug reports contain useful information. However, searching for specific content is difficult because there are many bug reports. The desired content depends on the viewpoint of the stakeholder. A full text search includes unwanted content, which is costly. Although this problem has been previously noted, a solution has yet to be proposed. Herein we propose Content-based Labeling Method as a solution. This method organizes information in a bug report by labeling each sentence based on its contents, allowing stakeholders’ viewpoints to be considered. We evaluate the improvement in searchability. The Content-based Labeling Method improves the searchability according to the F-measure and precision of the experimental results.

    AB - Most stakeholders refer to past bug reports when they encounter a problem since bug reports contain useful information. However, searching for specific content is difficult because there are many bug reports. The desired content depends on the viewpoint of the stakeholder. A full text search includes unwanted content, which is costly. Although this problem has been previously noted, a solution has yet to be proposed. Herein we propose Content-based Labeling Method as a solution. This method organizes information in a bug report by labeling each sentence based on its contents, allowing stakeholders’ viewpoints to be considered. We evaluate the improvement in searchability. The Content-based Labeling Method improves the searchability according to the F-measure and precision of the experimental results.

    KW - Bug report

    KW - Labeling

    KW - Machine learning

    KW - Searchability

    UR - http://www.scopus.com/inward/record.url?scp=85051822979&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85051822979&partnerID=8YFLogxK

    U2 - 10.1007/978-3-319-97679-2_8

    DO - 10.1007/978-3-319-97679-2_8

    M3 - Conference contribution

    SN - 9783319976785

    T3 - Smart Innovation, Systems and Technologies

    SP - 75

    EP - 85

    BT - Knowledge-Based Software Engineering

    A2 - Kumeno, Fumihiro

    A2 - Oikonomou, Konstantinos

    A2 - Virvou, Maria

    PB - Springer Science and Business Media Deutschland GmbH

    ER -