Anticipating Bugs: Ticket-Level Bug Prediction and Temporal Proximity Effects

2025-06-17

AI Summary

Traditional defect prediction focuses solely on already-defective code segments and supports only post-hoc repair, not prevention. To address this, we propose Ticket-Level Defect Prediction (TLP), the first approach to shift defect prediction earlier into the ticket lifecycle—specifically to the Open, In Progress, and Closed stages—enabling proactive defect prevention. TLP integrates 72 heterogeneous features spanning code attributes, developer profiles, internal/external “temperature” (activity trends), intrinsic complexity, ticket-code associations, and Just-In-Time (JIT) commit signals, leveraging sliding windows and three machine learning classifiers. Key contributions include: (1) pioneering defect prediction at the requirement/task level; (2) revealing that feature effectiveness dynamically evolves with ticket progression, with temporal proximity significantly improving prediction accuracy; and (3) empirical validation on >10,000 Apache tickets showing steadily increasing accuracy as tickets advance—developer features dominate early-stage prediction, code/JIT features peak at closure, and temperature features consistently deliver complementary gains throughout.

Technology Category

• Machine Learning: Feature Construction/Reformulation
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Planning, Routing, and Scheduling: Temporal Planning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

The primary goal of bug prediction is to optimize testing efforts by focusing on software fragments, i.e., classes, methods, commits (JIT), or lines of code, most likely to be buggy. However, these predicted fragments already contain bugs. Thus, the current bug prediction approaches support fixing rather than prevention. The aim of this paper is to introduce and evaluate Ticket-Level Prediction (TLP), an approach to identify tickets that will introduce bugs once implemented. We analyze TLP at three temporal points, each point represents a ticket lifecycle stage: Open, In Progress, or Closed. We conjecture that: (1) TLP accuracy increases as tickets progress towards the closed stage due to improved feature reliability over time, and (2) the predictive power of features changes across these temporal points. Our TLP approach leverages 72 features belonging to six different families: code, developer, external temperature, internal temperature, intrinsic, ticket to tickets, and JIT. Our TLP evaluation uses a sliding-window approach, balancing feature selection and three machine-learning bug prediction classifiers on about 10,000 tickets of two Apache open-source projects. Our results show that TLP accuracy increases with proximity, confirming the expected trade-off between early prediction and accuracy. Regarding the prediction power of feature families, no single feature family dominates across stages; developer-centric signals are most informative early, whereas code and JIT metrics prevail near closure, and temperature-based features provide complementary value throughout. Our findings complement and extend the literature on bug prediction at the class, method, or commit level by showing that defect prediction can be effectively moved upstream, offering opportunities for risk-aware ticket triaging and developer assignment before any code is written.

Problem

• Predicts tickets likely to introduce bugs post-implementation
• Analyzes feature reliability across ticket lifecycle stages
• Identifies shifting predictive power of feature families temporally

Innovation

• Ticket-Level Prediction (TLP) for bug anticipation
• 72 diverse features across six families
• Sliding-window ML approach on 10,000 tickets