SBIR Phase I: Artificial Intelligence (AI) System for Enterprise Software Incident Management

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project will be improving the robustness and performance of enterprise information infrastructure. Automated solutions for incident management are critical. The proposed project develops an automated artificial intelligence (AI) system to interpret reports of IT problems and recommend solutions. This Small Business Innovation Research (SBIR) Phase I project aims at transforming IT incident management for enterprise companies. The novelty of the proposed solution lies in two unique features: 1) intelligent learning techniques that can automatically infer effective remediation actions without requiring tedious and error-prone user-defined rules; and 2) active learning capabilities which can adaptively refine the prevention actions based on continuous system health monitoring data and user feedback. Specifically, the proposed project consists of three thrusts: 1) building universal pattern extraction schemes which can extract precise and reusable patterns from various data sources including complex incident ticket data, metric data, and log data; 2) exploring a self-learning remediation recommendation system that can recommend proper remediation by analyzing both the incident pattern and root cause patterns; and 3) easy-to-use remediation workflow engine which allows the user to trigger different remediation workflows to prevent the predicted incident in one place. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.