Large-Scale Disk Failure Prediction

This book constitutes the thoroughly refereed post-competition proceedings of the AI Ops Competition on Large-Scale Disk Failure Prediction, conducted between February 7th and May 15, 2020 on the Alibaba Cloud Tianchi Platform. A dedicated workshop, featuring the best performing teams of the competition, was held at the 24th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2020, in Singapore, in April 2019. Due to the COVID-19 pandemic, the workshop was hosted online.

This book includes 13 selected contributions: an introduction to dataset, selected approaches of the competing teams and the competition summary, describing the competition task, practical challenges, evaluation metrics, etc.

Inhalt
An Introduction to PAKDD CUP 2020 Dataset.- PAKDD 2020 Alibaba AIOps Competition-Large-scale Disk Failure Prediction: Third Place Team.- A Voting-based Robust Model for Disk Failure Prediction.- First place solution of PAKDD Cup 2020.- Anomaly Detection of Hard Disk Drives based on Multi-scale Feature.- Disk Failure Prediction: an in-depth comparison between deep neural networks and tree-based models.- PAKDD2020 Alibaba AI Ops Competition: Large-Scale Disk Failure Prediction.- SHARP: SMART HDD Anomaly Risk Prediction.- Tree-based model with advanced data preprocessing for large scale hard disk failure prediction.- PAKDD2020 Alibaba AI Ops Competition: An SPE-LightGBM Approach.- Noise Feature Selection Method in PAKDD 2020 Alibaba AI Ops Competition: Large-Scale Disk Failure Prediction.- Characterizing and Modeling for Proactive Disk Failure Prediction to Improve Reliability of Data Centers.- Summary of PAKDD CUP 2020: From Organizers' Perspective.