Maestro: A Deep Learning Based Tool to Find and Explore Architectural Design Decisions in Issue Tracking Systems

Jesse Maarleveld*, Arjan Dekker, Sarah Druyts, Mohamed Soliman

*Corresponding author voor dit werk

OnderzoeksoutputAcademicpeer review

2 Citaten (Scopus)
61 Downloads (Pure)

Samenvatting

Software engineers commonly re-use architectural design decisions (ADDs) from their previous experience. However, in practice, software engineers still depend on adhoc mechanisms to re-use ADDs. Recent studies show that software engineers discuss ADDs in issue tracking system, which could be useful for software engineers to make new ADDs. Nevertheless, it is rather challenging to find ADDs among the big amount of issues in issue trackers. Therefore, we introduce Maestro, an open source tool for finding, annotating, and exploring ADDs in issue tracking systems. The tool allows researchers and practitioners to find and analyze issues containing ADDs in issue trackers. Maestro provides annotation mechanisms, deep learning components, keywords-based search engine and a user-interface that can be easily used by researchers and practitioners to find and analyze ADDs in issue trackers.

Originele taal-2English
TitelSoftware Architecture
SubtitelECSA 2023 Tracks, Workshops, and Doctoral Symposium, Revised Selected Papers
RedacteurenBedir Tekinerdoğan, Romina Spalazzese, Hasan Sözer, Silvia Bonfanti, Danny Weyns
UitgeverijSpringer
Pagina's390-405
Aantal pagina's16
ISBN van elektronische versie978-3-031-66326-0
ISBN van geprinte versie978-3-031-66325-3
DOI's
StatusPublished - 2024
EvenementProceedings of the 17th European Conference on Software Architecture, ECSA 2023 - Istanbul, Turkey
Duur: 18-sep.-202322-sep.-2023

Publicatie series

NaamLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
UitgeverijSpringer
Volume14590 LNCS
ISSN van geprinte versie0302-9743
ISSN van elektronische versie1611-3349

Conference

ConferenceProceedings of the 17th European Conference on Software Architecture, ECSA 2023
Land/RegioTurkey
StadIstanbul
Periode18/09/202322/09/2023

Vingerafdruk

Duik in de onderzoeksthema's van 'Maestro: A Deep Learning Based Tool to Find and Explore Architectural Design Decisions in Issue Tracking Systems'. Samen vormen ze een unieke vingerafdruk.

Citeer dit