Unsupervised Anomaly Detection in Cloud-Native Microservices via Cross-Service Temporal Contrastive Learning: Risk Stratification and Threshold Selection

Abstract

Risk stratification is a decision problem in which thresholds, class prevalence, error costs, and downstream actions must be evaluated together. This structured evidence review evaluates "Unsupervised Anomaly Detection in Cloud-Native Microservices via Cross-Service Temporal Contrastive Learning" alongside nine author-disjoint, topically matched publications in cloud-native anomaly detection. It compares construct definitions, evaluation choices, operating assumptions, and reported limitations instead of treating bibliographic similarity as empirical equivalence. Viewed through risk stratification and threshold selection, the map separates claims supported by the available record from questions that still require full-text extraction, replication, or new experiments. The synthesis is interpretive rather than meta-analytic and therefore does not present a pooled effect estimate or a new causal result. The resulting agenda compares thresholds under observed prevalence, quantifies error costs, and links each stratum to a defined action.