iSPY: Issue-Solution Pairing From Developer Live Chats

iSPY: Issue-Solution Pairing From Developer Live Chats

Live chat streams are often used in open source software (OSS) development as a fast way to share information around the problem. But when live chat gets used a lot, important project information is lost in the middle of so many streams. AI to rescue! Here we describe “iSPY”, a scalable tool for hel...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE software 2022-11, Vol.39 (6), p.109-113
Hauptverfasser: Shi, Lin, Yang, Ye, Wang, Qing, Menzies, Tim
Format: Artikel
Sprache:eng
Schlagworte:
Artificial intelligence
Neural networks
Open source software
Streams
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Live chat streams are often used in open source software (OSS) development as a fast way to share information around the problem. But when live chat gets used a lot, important project information is lost in the middle of so many streams. AI to rescue! Here we describe “iSPY”, a scalable tool for helping developers understand busy live chat streams. In a study with 173,278 live chats, we show that iSPY can automatically extract issue-solution pairs from intensive developer live-chats (using neural networks technology). In this article we introduce iSPY as well as discussing future directions for learning knowledge on issue resolution from crowd developers.
ISSN:0740-7459
1937-4194
DOI:10.1109/MS.2022.3196695