RAG4Tickets: AI-Powered Ticket Resolution via Retrieval-Augmented Generation on JIRA and GitHub Data

2025-10-09

AI Summary

Modern software teams face delayed issue resolution due to fragmented knowledge scattered across JIRA tickets, GitHub pull requests, and developer discussions. To address this, we propose a cross-platform knowledge fusion method integrating JIRA and GitHub data. Our approach establishes a unified data processing pipeline and introduces a semantic embedding and hierarchical indexing strategy for heterogeneous artifacts—leveraging Sentence-Transformers for embedding and FAISS for efficient retrieval. Furthermore, we design an evidence-driven repair generation module based on a retrieval-augmented generation (RAG) framework, employing large language models to produce interpretable, context-aware fix suggestions. Evaluated on real-world projects, our system improves ticket resolution accuracy by 18.7% over baseline methods, reduces average resolution time by 32%, and achieves an 89.4% adoption rate for generated fixes. These results demonstrate significant gains in knowledge reuse efficiency and explainable recommendation capability within DevOps environments.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Modern software teams frequently encounter delays in resolving recurring or related issues due to fragmented knowledge scattered across JIRA tickets, developer discussions, and GitHub pull requests (PRs). To address this challenge, we propose a Retrieval-Augmented Generation (RAG) framework that integrates Sentence-Transformers for semantic embeddings with FAISS-based vector search to deliver context-aware ticket resolution recommendations. The approach embeds historical JIRA tickets, user comments, and linked PR metadata to retrieve semantically similar past cases, which are then synthesized by a Large Language Model (LLM) into grounded and explainable resolution suggestions. The framework contributes a unified pipeline linking JIRA and GitHub data, an embedding and FAISS indexing strategy for heterogeneous software artifacts, and a resolution generation module guided by retrieved evidence. Experimental evaluation using precision, recall, resolution time reduction, and developer acceptance metrics shows that the proposed system significantly improves resolution accuracy, fix quality, and knowledge reuse in modern DevOps environments.

Problem

• Addresses fragmented knowledge across JIRA and GitHub delaying issue resolution
• Integrates semantic embeddings with vector search for context-aware recommendations
• Generates explainable resolution suggestions by synthesizing historical ticket data

Innovation

• RAG framework integrates Sentence-Transformers and FAISS search
• Embeds JIRA tickets, comments, and GitHub PR metadata
• LLM synthesizes retrieved cases into resolution suggestions