Gold Partner of HUSTEF 2019: Huba Demeter, Nokia
summary
Nokia applies unsupervised neural sequence analytics to detect anomalies during software upgrades, trained only on green reference data from successful upgrades. A general log parsing engine ingests diverse logs from network functions that handle millions of calls and must meet five‑nines availability. The system highlights the few anomalous sequences so engineers can decide within tight maintenance windows, reducing manual log triage and surfacing upgrade failures earlier than rule‑based checks.
AI-written summary · about
related talks
29mEnhancing Business Process Anomaly Detection With Neo4j and Graph Neural Networks
47m Natural Language Processing & Anomaly detection in Sys call logs
29mAnomaly Detection
18mProduct Reliability in
30mHitachi applying AI in Industry 4.0 digital transformation projects
48mUtilizing Lithuanian Public Sector Big Data to Make the Country Run Smarter
21mDebugging at Scale—Going from Single Box to Production
22mEngineering Predictive Platforms