Towards a Newcomers Dataset to Assess Conversational Agent�s Efficacy in Mentoring Newcomers
Many Open Source Software (OSS) maintainers experience burnout due to the constant need to support and answer questions from newcomers. This leads to slower response times and unanswered queries, discouraging newcomers and ultimately hindering community growth. Research indicates that conversational agents trained on community data can answer common questions, thus reducing the workload for maintainers. However, developers require real questions from newcomers extracted from an actual project to test the efficacy and guide the development of such chatbots. To support the development of conversational agents for newcomer onboarding, we curated a dataset of varied questions from the JabRef OSS project. This dataset paper details our collection process, describes the data, and offers reflections on future research. We expect that this dataset can foster further research focused on developing chatbots to assist newcomers.
Sun 27 AprDisplayed time zone: Eastern Time (US & Canada) change
16:00 - 17:30 | Session 3: Evaluating and Improving Bot ImpactBotSE at 213 Chair(s): Ahmad Abdellatif University of Calgary Ahmad Abdellatif | ||
16:00 22mTalk | Towards a Newcomers Dataset to Assess Conversational Agent�s Efficacy in Mentoring Newcomers BotSE Misan Etchie NAU RESHAPE LAB, Hunter Beach NAU RESHAPE LAB, Katia Romero Felizardo NAU RESHAPE LAB, Igor Steinmacher NAU RESHAPE LAB | ||
16:22 22mTalk | Bot-Driven Development: From Simple Automation to Autonomous Software Development Bots BotSE Link to publication DOI Pre-print | ||
16:45 22mTalk | Bridging HCI and AI Research for the Evaluation of Conversational SE Assistants BotSE | ||
17:07 22mTalk | Reducing Alert Fatigue via AI-Assisted Negotiation: A Case for Dependabot BotSE Raula Gaikovina Kula The University of Osaka |