KabOOM: Unsupervised Crash Categorization through Timeseries Fingerprinting

TL;DR

KabOOM is an unsupervised method that uses timeseries fingerprinting with auto-encoders to automatically cluster mobile app crash reports into meaningful categories, aiding developers in root cause analysis.

Contribution

Introducing KabOOM, a novel unsupervised approach leveraging multivariate timeseries fingerprinting and auto-encoders for effective crash categorization in mobile applications.

Findings

Successfully clusters crash reports into intuitive categories

Reduces dimensionality of telemetry data

Effective in aiding crash root cause analysis

Abstract

Modern mobile applications include instrumentation that sample internal application metrics at regular intervals. Following a crash, sample metrics are collected and can potentially be valuable for root-causing difficult to diagnose crashes. However, the fine-grained nature and overwhelming wealth of available application metrics, coupled with frequent application updates, renders their use for root-causing crashes extremely difficult. We propose KabOOM, a method to automatically cluster telemetry reports in intuitive, distinct crash categories. Uniquely, KabOOM relies on multivariate timeseries fingerprinting; an auto-encoder coupled with a cluster centroid optimization technique learns embeddings of each crash report, which are then used to cluster metric timeseries based crash reports. We demonstrate the effectiveness of KabOOM on both reducing the dimensionality of the incoming crash reports and producing crash categories that are intuitive to developers.